Professional Documents
Culture Documents
(ICMFE-2014)
Proceedings
3rd International Conference on Management, Finance and
Entrepreneurship
Penang, Malaysia
13 -14, December 2014
ISSN 2311-6269
Organized by
Preface
Dear Distinguished Delegates and Guests,
The Conference Committee warmly welcomes our distinguished delegates and guests to the 2014 International
Conference on Management, Finance and Entrepreneurship (ICMFE-2014) held on December 13-14 in Penang,
Malaysia.
The ICMFE-2014 is organized by International Foundation for Research and Development (IFRD). The
conference is aimed at discussing with all of you the wide range of problems encountered in present and
future issues in economies and Societies. The ICMFE-2014 is organized in collaboration with Yildirim Beyazit
University, Turkey, Shinawatra International University, Thailand, PERTRE ANDERI of IASI, Romania and
National Academy of Management, Ukraine where researchers from around the world presented their work.
The conference committee is itself quite diverse and truly international, with membership around the world.
The proceeding records the fully refereed papers presented at the conference. The main conference themes
and tracks are Management, Finance and Entrepreneurship. The conference aims to bring together
researchers, scientists, engineers and practitioners to exchange and share their experiences, new ideas and
research results about all aspects of the main conference themes and tracks and discuss the practical
challenges encountered and the solutions adopted. The main goal of the event is to provide a scientific forum
for exchange of new ideas in a number of fields that interact in depth through discussions with their peers
from around the world.
The conference has solicited and gathered technical research submission related to all aspects of major
conference themes and tracks. All the submitted papers have been peer reviewed by the reviewers drawn
from the scientific committee, external reviewers and editorial board depending on the subject matter of the
paper. Reviewing and initial selection were undertaken electronically. After the rigorous peer-review process,
the submitted papers were selected based on originality, significance, and clarity for the purpose of the
conference. The conference program is extremely rich, featuring high-impact presentations. The high quality
of the program guaranteed by the presence of an unparalleled number of internationally recognized top
experts. The conference will therefore be a unique event, where attendees will be able to appreciate the latest
results in their field of expertise, and to acquire additional knowledge in other fields. The program has been
strutted to favor interactions among attendees coming from many diverse horizons, scientifically,
geographically, from academia and from industry.
We would like to thank the program chairs, organization staff, and members of the program committee for
their work. We are grateful to all those who have contributed to the success of ICMFE-2014 especially our
partners. We hope that all participants and other interested readers benefit scientifically from the
proceedings and find it stimulating in the process. Finally, we would like to wish you success in your technical
presentations and social networking.
We hope you have a unique, rewarding and enjoyable time at ICMFE-2014 in Penang.
With our warmest regards,
Conference Committee
December 1314, 2014
Penang, Malaysia.
ii
ICMFE-2014
Conference Committee
Conference Chair
Dileep Kumar, M., Ph. D., University Utara Malaysia, Malaysia
Conference Convener
Nek Kamal Yeop Yunus, Ph. D., Univerisiti Pendidikan Sultan Idris , Perak, Malaysia
Members
John Walsh, Ph. D., Shinawatra International University, Bangkok, Thailand
Sisira R. N. Colombage, Ph. D., Monash University, Victoria, Australia
Nek Kamal Yeop Yunus, Ph. D., Univerisiti Pendidikan Sultan Idris , Perak, Malaysia
R K Uppal, Ph. D., DAV College, Punjab, India
Rishidaw Balkaran, Ph. D., Durban University of Technology, South Africa
Ayhan Kapusuzoglu, Ph. D., Yildirim Beyazit University, Turkey
M. Saman Dassanayake, Ph. D., University of Colombo, Colombo, Sri Lanka
Katalin Jackel, Ph. D., Budapest Business School, Budapest, Hungary
Wei-Bin Zhang, Ph. D., Ritsumeikan Asia Pacific University, Japan
Susantha Herath, Ph. D., St. Cloud State University, USA
Chandana Prasad Withana, Ph. D., Charles Sturt University, Sydney, Australia
Chux Gervase Iwu, Ph. D., Cape Peninsula University of Technology, South Africa
Somnath Sen, Ph. D., University of Birmingham, United Kingdom
Johan de Jager, Ph. D., Tshwane University of Technology, South Africa
Kevin Feeney, Ph. D., American University in Bulgaria
Ahasanul Haque, Ph. D., International Islamic University Malaysia (IIUM), Malaysia
Izah Mohd Tahir, Ph. D., University Sultan Zainal Abidin, Terengganu, Malaysia
Pratibha Samson Gaikwad, Ph. D, Shivaji University of Pune, India
iii
Table of Contents
Description
Pages
Preface
Ii
Conference Committee
Table of Contents
iii
Papers
vi
iv
iv
1
13
15
23
36
43
54
60
68
72
80
89
94
103
111
119
126
132
143
149
156
PAPERS
vi
Retailer
Sell
Buy
components
and
materials
Manufacture
Deliver
This responsive business model is the theory that then being employed by researchers to determine the
theoretical framework. The manufacturing factors are the aspect inside the manufacture, and how it will
affect the fourth aspect of this theory which is delivery which will be represented by on-time delivery. The
responsive business model is a narrower concept of supply chain exposure and performance. Therefore,
researchersare using this model as the theoretical framework. However, they will also facing the supply chain
risk or exposure in the global supply chain.In this research, researchers would like examine the factors in
manufacturing activities and how it affects the supply chain performance that is represented by the on-time
delivery in LG Indonesia.
Statement of the Problem: Based on some discussions that have been made by researchers and LG supply
chain stakeholders, researchers identified three main manufacturing factors that could affected the
performance of the supply chain in LG which are shipment plan change rate (SPCR), production observant
rate (POR), and also production change rate (PCR). Hence, the paper will examine how those manufacturing
factors will affect the performance of the supply chain in LG Indonesia. Because of the problem that has been
identified in the problem identification, the research question that needs to be answered is;
Is there any impact of manufacturing factors towards on time delivery in PT LG Elektronik
Indonesia?
Research Objective: The objective of this research based on above mentioned research question are;
a. To examine the impactof Shipment Plan Change Rate (SPCR) to the On -Time Delivery (OTD).
b. To examine the impact ofProduction Observant Rate (POR) to the On-Time Delivery (OTD).
c. To examine the impact of Production Change Rate (PCR) to the On-Time Delivery (OTD).
2. Literature Review
Supply Chain Management: Siagian (2007) stated that supply chain involved the whole interaction among
suppliers, distributors, and customers. This integration could occur in the form of transportation,
information, scheduling, and even the transfer of raw materials among involved parties.According to
Rahmasari (2011), every company in the supply chain has the purpose to satisfy the customer by creating a
product with low-cost, on-time delivery, and good quality. Those purposes can be accomplished by measuring
the performance of the supply chain management with the following measurements:
1. Quality (customer satisfaction, customer loyalty, on-time delivery)
2. Time ( total replenishment time, business cycle time)
3. Cost (total delivery cost, value-added efficiency)
4. Flexibility (amount and specifications)
Supply Chain Exposure: According to Manuj and Mentzer (2008), the risk or exposure in the supply chain
comes from three sources. Those three sources are demand exposure, supply exposure, and operational
exposure.
Demand Exposure: Demand exposure relates to potential or actual disturbances to flow of product,
information, and cash, emanating from within the network, between the focal company and the market. This
SPCR (X1)
H1
H2
POR (X2)
OTD (Y)
H3
PCR (X3)
H4
Manufacturing Factors
Delivery
3. Methodology
Operational Definitions
a.
b.
c.
d.
Shipment Plan Change Rate (SPCR): Shipment Plan Change Rate is a tool that being occupied in PT LG
Elektronik Indonesia to measure the ratio changes between shipment plan and the actual shipment. The
unit for this measurement is percentage.
Production Observant Rate (POR): Production Observant rate is one of the parts in the production plan
in LG. In the production plan, POR is used to measure the accuracy of the production plan that has been
developed. Production Observant Rate is a tool that being occupied in PT LG Elektronik Indonesia to
measure the ratio between the production plan and the actual production. The unit for this measurement
is percentage.
Production Change Rate (PCR): Production Change rate is one of the parts in the production plan in LG.
In the production plan, PCR is used to measure the accuracy of the production plan that has been
developed. The Production Change Rate is a tool that used in PT LG Elektronik Indonesia to measure the
ratio of changes in customer order. The unit for this measurement is percentage.
On-Time Delivery: On time delivery is one of the key performance indicator of the supply chain in PT LG
Elektronik Indonesia. In some literatures, on time delivery also considered as the supply chain reliability.
The unit for this measurement is percentage.
Research Design: The method that researchers use in order to conduct this thesis is quantitative analysis by
using secondary data in which researchers will give some statistical treatment to all of the data got from the
PT LG Elektronik Indonesia. Researchers will also using the deductive approach in which researchers develop
a hypothesis in the beginning and proof it by a series of calculation.
Hypotheses
H1: Shipment Plan Change Rate (SPCR) has a significant impact towards On Time Delivery
H2: Production Observant Rate (POR) has a significant impact towards On Time Delivery
H3: Production Change Rate (PCR) has a significant impact towards On Time Delivery
H4: Shipment Plan Change Rate (SPCR), Production Observant Rate (POR), and Production Change Rate
(PCR), has a significant impact toward On-Time Delivery.
Production Observant
Rate (POR)
Production
Rate (PCR)
W1
W2
W3
W4
W5
W6
W7*
W8*
W9*
W10
W11
W12
W13
W14
W15
0.00%
0.31%
2.61%
1.80%
1.90%
0.17%
0.03%
0.21%
0.08%
1.71%
0.57%
2.08%
1.58%
1.88%
0.92%
16.74%
7.22%
4.39%
5.99%
9.01%
9.85%
178.52%
74.74%
71.10%
22.05%
8.23%
8.91%
10.68%
8.26%
11.87%
0.00%
0.00%
0.00%
0.38%
0.05%
0.03%
100.00%
76.62%
7.69%
0.18%
0.04%
0.00%
0.20%
0.00%
0.00%
Change
On Time Delivery
(OTD)
89.45%
96.21%
94.90%
94.26%
94.53%
90.33%
85.42%
84.68%
85.57%
88.05%
83.90%
89.14%
93.96%
89.18%
85.39%
10
11
SD
5.60
10.14
6.15
with
life.
R2
Step 1
.05
Age
.02
Gender
-.03
Marital status
.15
Sect
.09
Step 2
.23***
Religious Rituals
.37**
Religious Muamalats
.21*
2
Total R
.29
Note. *p < .05.**p < .01. ***p < .001
Hierarchal Multiple Regression Analysis was carried out to find out the predictors of life satisfaction. In first
step life satisfaction was entered as outcome variable and demographics (age, gender, sect and marital status)
were entered as predictors. No significant predictor emerged in this step. In second step religious rituals and
maumalats were entered as predictors and life satisfaction as outcome variable. Religious rituals appeared as
a significant predictor of life satisfaction with = .37, p< .01. Religious maumalats also emerged as a
significant predictor of life satisfaction with = .21, p <.05. Demographics (age, gender, sect and marital
status) and religiosity overall, accounts 29% variance on life satisfaction.
Discussion: The current study was conducted to find out the associationbetween life satisfaction and
religiosity among college teachers. The sample of the study was teachers of colleges of district Gujarat.
Responses from 100 teachers provide the basics following findings. Results of current study are also in line
with those of conducted in other Islamic countries.Khalek& Lester (2013), Khalek (2010) studied the positive
relationship between subjective well-being and religiosity among Kuwaiti undergraduates.Aghili& Kumar
(2008) studied the happiness feelings in Iranian employees and found positive relationship between
religiosity and happiness.Tiliouine&Belgoumidi (2008) investigated the positive relationship between
religiosity and satisfaction with life in Muslim students from Algeria. Current results that religiosity and life
satisfaction are positively related to each other are also consistent with studies conducted in non-Muslim
countries. Barbera&Grhan (1997) conducted a study to understand the positive relationship of life
satisfaction with selected secular and sacred values in New York. Lima and Putnam (2010) conducted study
in America and provide strong evidence of impact of religion on life satisfaction. Levin (2011) studied positive
relationship between satisfaction with life and importance of Gods in life, in samples of Jewish respondents
from Israel. Han & Lai (2012) studied the positive association between religion and life satisfaction in
convenient samples of Chinese in Hong Kong and Australia.
In many researches similar findings found with different scales to measure life satisfaction and religiosity.Gull
&Dawood (2013) used Religiosity Index and Trait Well-being Inventory to study the relationship between life
satisfaction and religiosity. Khalek& Lester (2013) investigated positive association between life satisfaction
and religiosity, usingArabic Scale of Mental Health (ASMH) and self-rating scales to assess physical health,
psychological health, and religiosity. Khalek (2010) used World Health Organization QOL scale-Brief
(WHOQOL-Bref) and six self-rating scales of physical health, mental health, happiness, satisfaction with life
and religiosity to investigate the relationship between life satisfaction and religiosity. The findings of current
study suggested that there is a significant positive relationship between life satisfaction and religiosity. So, it
supports the hypothesis of the study that life satisfaction will be positively related to religiosity. These
findings are consistent with those of Baco, 2010; Barbera&Gurhan, 1997; Chang, 2009; Gull &Dawood, 2013;
Han & Lai, 2012; Khalek, 2010; Khalek& Lester, 2013; Levin, 2011; Lima &Putnum,
2010;Mccullough&Willough, 2009; Tiliouine&Belgoumidi, 2008. Findings of their studies revealed that there
is a significant positive relationship between life satisfaction and religiosity.
12
13
14
15
16
Attributes
Production
Process
project
depend
on
depend
on
Economical Factors: In this process we have to score Economical aspects of the project. To do this we need
to fill the table 2. The aim is to find out the benefits, the accuracy of financial planning in a project and finding
sensitivity of the project. For instance assuming the costs of a project more than the reality have negative
point and high benefit and margin for the project are positive points.
17
Attributes
To what extent do the project costs comply with the
requested investment amount?
To what extent does the breakeven point proper in
this project?
To what extent the benefit in this project is
appropriate?
Attributes
18
Attributes
Bargaining
Power
of
Suppliers
Bargaining
Power
of
Buyers
Threat
of
new
Entrants
Substitute
Products
Strategic Factors: As the title indicates in this process we have to score strategic aspects of the project. To do
this we need the research and data gathering to score following table 5. The aim is to find out if a project is
aligned with the company mission. For instance compatibility of the project with PND and country strategies
could be considered as a positive point for the project.
19
Attributes
Weight
Technical factors
Economical factors
Risk factors
Market factors
Strategic factors
Social factors
3. Conclusion
In this paper we have attempted to develop a scoring model which can be used for venture capitalists for
initial evaluation of the projects. We began by reviewing the literature in relevant papers and books. We have
20
21
22
Sharia Based (Islamic Banking) Financing and its Relevance to Sultanate of Oman
Yasmeen Ali Al-Balushi
Rustaq College of Applied Sciences, Ministry of Higher Education, Sultanate of Oman
yasmeen.rus@cas.edu.om
Abstract: In modern days, not only traditional financing gaining importance but also the interest free
microfinance gaining an enormous importance to micro entrepreneurs whether it is in developing or least
developing countries. This type of financing system provide loan to its clients without interest and there are
many different investment modes of funding. This paper present the general trends to the people in showing
their interest in Shariah based financing in the sultanate of Oman and how the institutional credit facilities
may reach to the rural-based small entrepreneurs by Islamic banking finance and their relationships. Further,
a qualitative nature of research methodology is used in this paper; qualitative research enables the
researcher to know in depth the aspects of a case and it provides all the facilities in getting and collecting the
information. In order to see the reality of the demand of Islamic Banking in the environment of the Sultanate a
depth interviews were managed with a group of people. The main respondents in the present study were
group of banks employees form Islamic banks and Islamic windows of conventional banks. The end
results of the paper were that there is huge demand of Islamic banking system among Omani people and it
also shows how Islamic banking system would empower small entrepreneurs.
Keywords: Islamic Banking, finance, banking, micro entrepreneurs
1. Introduction
Islamic banking finance is based on Shariah (Islamic Law), the lending and borrowing modes of which are
different from a traditional bank. Islamic banks lend funds to their customers in kind rather than in cash and
charge profits on their borrowed funds. Money in Islamic finance is not a commodity but only a mean to settle
a transaction. Islamic banking is relatively recent development and many Islamic countries have realized that
dealing with western banking and finance methods are undermining their adherence to Islamic principles
(Taylor, 2003). Therefore many Islamic countries reform their financial practice. Islamic finance offers a great
incentive for Islamic countries to develop financial market (Ibrahim, 2008). There were at least about 176
Islamic banks in the early 2003 around the world and about 32 of them are operating in an Arab state
(Ghannadian & Goswami, 2004). Trade and commerce are not new to the Islam the Prophet Mohammed was
a merchant and a trader (Ibrahim, 2008). Today and in the Muslim history Islamic entrepreneurs always find
the opportunities to commerce and trade which satisfy them to the Islamic beliefs (ibid). There are many
sources of Sharia. The primary source like Holy Quran, the Hadith (saying of the Prophet) and the Sunnah
(practices of the Prophet) are the primary source of Islamic law and principal guide. Holy Quran is the first
revealed source. Qiyas (analytical deduction and reasoning), Ijma (consensus of Sharia scholars) and Ijtihad
(legal reasoning) are the secondary sources of sharia. In addition, further sources in Islam are Istihsan
(personal preference) Istislah (public welfare), Darurah (necessity) and Urf (custom).
Oman is one of the largest countries in the Gulf region with population of 3,113000
(www.moneoman.gov.om). In the seventh century Oman has adopted the Islam during the life time of
Prophet Muhammad. (Background Note: Sultanate of Oman, 2004). Being an Islamic country presently
Omans government has established within last two years the legal and regulatory framework which applies
to Islamic banks and Islamic windows of conventional banks . The Sultanate Oman was the last country in
the six-nation Gulf cooperation Council to introduce Islamic finance . On other hand, all GCC countries have
Islamic bank and there are more than 20 Islamic Banks in the GCC. For example, Saudi Arabia (Islamic
Development Bank), Kuwait (Kuwait International Bank), Bahrain (Bahrain Islamic Bank), Qatar (Qatar
Islamic Bank), and The United Arab Emirates (Dubai Islamic Bank).This motivates me to do my dissertation
about Islamic Banking and it is relevant in the context of the economy of Sultanate of Oman. According to the
Finance of International Trade in the Gulf Arab States (1993) that during the period of 1920 and 1931 the
residents of the Gulf Arab States were disagreeing to deal with the conventional banks because the bank
activities were against their Islamic principle which were dealing with interest, once there were discovery of
oil in commercial quantities in the Gulf, many conventional banks want to take the opportunities to deal with
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
EDUCATIONAL
QUALIFICATION
LENGTH OF SERVICE
ON-JOB TRAINING
12-16
17-21
22-26
27-31
31and above
Non Formal Education
Primary Education
Secondary Education
OND
HND/B.Sc & Above
Below 5 years
6-15 years
16 years & above
Very Frequent
Regular
Occasional
Rare
10
3
13
34
57
16
26
5
8
62
18
75
24
66
28
12
11
117
117
117
117
8.54
2.56
11.11
29.05
48.71
13.67
22.22
4.28
6.84
52.99
15.39
64.10
20.51
56.41
23.93
10.26
9.40
%
100%
100%
100%
100%
100%
Interpretation: Table 1 reports the classification of respondent socio economic characteristics. The table
shows that 87.2% are male while 12.8% are female. It also revealed that respondents above the age of
27account for 77.76% of the entire survey population. More so, their qualification show that 52.99% of the
respondents have a Bachelors degree or HND and above. 64.10% has 6-15years length of service, 20.51% has
over 16years length of service while 15.39% has below 5years length of service. 56.41% and 23.93% have
very frequent and regular on-job the training while 10.26% and 9.4% are occasional and rare on the job
training.
To determine whether social infrastructure is a fundamental determinant of entrepreneurial
development: To test the stated hypothesis, the following were used as the independent variable (social
infrastructure) on entrepreneur performance (dependent) as stated in above: Hospitals/Health centres,
Schools, Water, Electricity, Transport, Market place & Housing.
Table 2: Showing Model Summary and Test of hypothesis of the fundamental determinant
relationship between Social Infrastructure and Entrepreneurial Development
Variables
R
R
Adjusted
Std. Error of F change Df1 Df2
Square
R
the Estimate
Social
Infrastructure
as .624 .68
.62
46483.362
4.3495E3 8
99
determinant of Entrepreneurial
Development
Source: Field Survey (2014)
Table 3: Regression result for the fundamental determinant relationship between Social
Infrastructure and Entrepreneurial development coefficientsa
Model
Standardized coefficients Beta
T
Sig.
1 (Constant)
Hospital/Health Centre (X1)
Schools (X2)
-2.138
-.446
-.085
-.055
-.011
39
.039
-.008
-.003
.208
1.784
.082*
Electricity (X4)
Transport (X5)
Market Place (X6)
Housing (X7)
.197
.417
.334
-.124
1.838
3.973
2.892
-.883
.073*
.000***
.007***
-.033
40
41
42
43
44
45
Code
Example
imx.y
Interview incident
ix.y
Category
Cx
c3: Category 3
Category property
Cpx
Insight
Nx
n9: Insight 9
Theoretical hypothesis
Thx
Following the above procedures, a theoretical model emerged. It was sealed when no major category
modifications were done with additional coding. As recommended by Glaser and Strauss (1967), this is the
point at which the theoretical model evolved into a sufficiently clear and logical form.
Table 2: Summary of Analytical Categories
Category group
Categories
TSB's work load (c6), TSB's equipment (c12), TSB's main business attraction
(c13), TSB's activities (c14), business reliability proofs (c15), perceived
weakness(es) (c21), managing weakness(es) (c22), following business standards
(c42), personal service (c49)
TSBs direct accessibility, basic price Companys office location (c4), Internet presence (c11), promotion (c24), pricing
and apprisal
(c26), marketing person (c32)
TSBs customers
Direct customers (c23), customer source market (c41), customer database (c43),
customers feedback/reactions (c47)
Origins (c19), good fortune (c20), business alliances (c25), relationship with
governmental institutions (c27), comparison to big businesses (c30), operational
difficulties (c34), number of employees (c37), business volume (c38), managing
seasonality (c39), formal business planning (c44), human resources (c45)
Environment
Competitive pressure (c28), issues with the government (c31), governments role
(c46)
Interview conditions
Transferability, Credibility, and Ethical Concerns: As this study is based on interpretive epistemology
rather than a positivist one, it is more suitable to look at transferability and credibility rather than reliability
and validity. Transferability and credibility for this study is built by using the comparative analysis technique,
a variety of respondents in the sample, and by the detailed presentation of procedures followed. To control
bias and subjectivity, systematic procedures and personal discipline were used (Glaser & Strauss, 1967;
46
Category Property
Illustrative Incident
c16
Knowledge
competition
of cp16.1
presence
Competitors' Ill tell you.. the camps available in, for example.. [one
competitors name].. ok.. ah.. there is.. there are three
ways to build [camping tents].. there is for example..
one of the camps.. he uses [details on competitors
tents] . . .
cp18.2
International NAUI is the oldest diving federation in the world,
associations
established by Jacques Cousteau.. and it is the oldest..
PADI was set up by people from NAUI.. and PADI is...
marketing [Laughing]!
cp18.5 Industry fragility
47
c26 Pricing
cp26.2
Matching
competition
c29
Understanding cp29.1 Core needs
customer needs
cp29.2 Secondary needs
(General
category,
category properties)
(General
category,
category properties)
(General
category,
category properties)
Insights: For this study, the grounded theory concept of insight was used to generate insightful ideas that
explain constructs in the emerging theory understood from the collective set of interviews and indicated by
the categories and category properties. Ten significant insights (out of fifteen) are presented below:
n2: Many satisfied customers become repeat customers. [Also indicated in cp23.6]
48
49
50
In addition, the model provides a methodology contribution by employing a simulated procedure of the
grounded theory technique of theoretical sampling. In essence, in cases where strict adherence to theoretical
sampling is unattainable, a simulating procedure could be used to achieve some of the sought benefits. Using
this procedure, the sample needs to be well planned and the order of analyzing collected data needs to be
carefully considered. The studys findings have also several practical contributions. They suggest that TSBs,
notwithstanding their inherent limitations, have a potential advantage if they focus on what they naturally do
best. Careful consideration of what market(s) to serve, coupled with active operational involvement and
exceptional personal service would lead to an increasingly better marketing performance. In essence, chances
of achieving good marketing are enhanced if the TSB owner enjoys working for his/her business. The study
evidently has a number of recommendations to researchers. The presented findings in the paper were the
most grounded on the data, but there were other glimpses of interesting areas to explore. First, a similar
approach on other sets of data in different regions would help build up the developed substantive theoretical
model into a formal one. In addition, direct analysis of customer perception of TSB marketing would be
helpful in testing the theoretical findings presented here, especially those pertaining to customer experience
elements.
References
Agndal, H., & Elbe, J. (2007). The Internationalization Processes of Small and Medium-sized Swedish Tourism
Firms. Scandinavian Journal of Hospitality and Tourism, 7(4), 301327.
Al-Azri, H. I., & Cai, L. (2009). Uniqueness of Tourism SMEs: Strategic Marketing Framework Propositions.
Tourism Issues, 8, 142151.
Al-Azri, H. I., & Morrison, A. M. (2006). Measurement of Oman's destination image in the US. Tourism
Recreation Research, 31(2), 8589.
Anderson, L. (1995). McCarthy's 4PS: Timeworn or Time-Tested? Journal of Marketing Theory and Practice,
3(3), 19.
Ateljevic, J. (2007). Small tourism firms and management practices in New Zealand: the centre stage macro
region. Tourism Management, 28, 307316.
Avcikurt, C., Altay, H., & Ilban, M. (2011). Critical Success Factors for Small Hotel Businesses in Turkey: An
Exploratory Study. Cornell Hospitality Quarterly, 52(2), 153164.
Buhalis, D. (2000). Marketing the competitive destination of the future. Tourism Management, 21(1), 97116.
Charmaz, K. (2004). Grounded Theory. In S. Hesse-Biber & P. Leavy, Approaches to qualitative research: a
reader on theory and practice (p. 496). New York, NY: Oxford University Press, Inc.
Etemad, H., & Wright, R. (2003). Globalization & entrepreneurship. Cheltenham, UK: Edward Elgar Publishing
Limited.
Fyall, A., & Garrod, B. (2005). Tourism Marketing: a collaborative approach. (C. Cooper, M. Hall, & D. Timothy).
Clevedon, UK: Channel View Publications.
Getz, D., & Carlsen, J. (2000). Characteristics and goals of family and owner-operated businesses in the rural
tourism and hospitality sectors. Tourism Management, 21, 547560.
51
52
53
54
55
1Thailands infrastructure is
incomplete
4. Results
Table 1 Thai Entrepreneurs Have Higher Operation Costs
Variables
1Thailand Infrastructure is Incomplete
2Fuel Prices Still Continuously Increase
3Thai Entrepreneurs Mainly Use Road Transportation
Total
*Number of respondents=50
56
Mean
2.94
2.90
2.84
2.89
S.D
0.24
0.56
0.25
0.35
Mean
3.04
S.D
0.25
2.79
2.58
0.37
0.42
2.8
0.34
Table 2 shows the mean and SD results for these variables: Thai Entrepreneurs Have Too Little
Understanding of Effective Transportation System; Thai Entrepreneurs Lack Knowledge and Skills about Just
in Time Concept; and Logistics Management is Separately Managed, the Collaboration in Each Department is
Low. The questionnaires results found that responses were in the agreed level in which the mean=2.80 and
SD=0.34. The mean of Thai Entrepreneurs Have Too Little Understanding of Effective Transportation System
was 3.04. The mean of Thai Entrepreneurs Lack Knowledge and Skills about Just in Time Concept was 2.79,
and the mean of Logistics Management is Separately Managed, the Collaboration in Each Department is Low
was 2.58. These three variables affect the ability of Thai entrepreneurs to effectively manage their inventory.
Operation costs and customer service were most affected by the entrepreneurs ability to understand the
principles of an effective transportation system. When there is too little understanding, costs rise and the
quality of customer service falls. The role that inventory plays in protecting a business from uncertainty has
been recognized for a long time. The disadvantage of keeping a big inventory for too long has also been
established, since this has more of an impact on inventory holding costs and it may affect managements
responsiveness (Baker, 2007). Water (2002) also reiterated that increased globalization leads to longer
supply lead times which requires a bigger inventory to satisfy customers. A further factor which affects the
transportation economy is the travel distance for big orders, since destinations that are far away require
more fuel, time and costs and this will increase the cycle stocks (Lawson, 2002). Currently the concepts of
continuous improvement and waste elimination are becoming better understood and are being integrated
into manufacturing practices. The competitive environment is affected by high demand for product reliability
and faster delivery (Robert and Rhonda, 2000). Another concept is the Just-in-Time system for waste
elimination that also reduces inventories at all levels, such as for raw materials, work-in-process and finished
goods inventories. Just-in-Time developed from a specific manufacturing practice in which suppliers of raw
materials eliminate waste and improve quality (Liker, 1998). Just-in-Time really needs support from all
functions for implementation to be successful. Lack of functional support, such as from finance, marketing
57
Mean
3.03
2.94
2.73
2.9
S.D
0.35
0.27
0.25
0.29
Table 3 shows the mean and SD results for these variables: Domestic Laws and Regulations are not
Competitive; the Cross-Border Trade Regulations Need to be Revamped and International Transportation
Rules Must be Changed. The results found that responses were in the agreed level in which the mean=2.90
and SD=0.29. The mean of Domestic Laws and Regulations are not Competitive, was 3.03. The mean of Crossborder Trade Regulations Need to be Revamped, was 2.94 and the mean of International Transportation
Rules Must be Changed, was 2.73. All the variables explain why Thai logistics rules and laws are not
competitive. Thai entrepreneurs agreed that the variable that increased operation costs and lowered
customer service the most was Domestic Laws and Regulations are not Competitive. Cambodias logistics
procedures are very complicated. For example, invoices and export certificates need to be submitted to the
Cambodian Government before any cargo can enter the country, which takes a long time (Asia News Monitor,
2014). Logistics procedures must reflect the changes in the business environment, because cargo transactions
have increased and the details are more complex. The Office of Transport and Traffic Policy and Planning
(OTP) was established to draft logistics management policies. It also works on transportation safety policies
but currently its role is limited to only policy creation for national and local governments (Asia News Monitor,
2013).
Table 4 Thai Logistics Workforce Lack Skills and Knowledge
Factors
1Thai Logistics Workforce Lack Skills and Logistics Management Experience
2 Thai Logistics Workforce Productivity is Lower than Entrepreneurs
Expectations
3 English Proficiency of Thai Logistics Workforce is Low
Total
*Number of respondents=50
Mean
3.45
2.97
S.D
0.54
0.37
2.78
3.06
0.29
0.4
Table 4 shows the mean and SD results for these variables: Thai Logistics Workforce Lack Skills and Logistics
Management Experience; Thai Logistics Workforce Productivity is Lower than Entrepreneurs Expectations;
English Proficiency of Thai Logistics Workforce is Low. The survey results showed that responses were in the
agreed level, in which the mean=3.06 and SD=0.4. The mean of Thai Logistics Workforce Lack Skills and
Logistics Management Experience, was 3.45. The mean of Thai Logistics Workforce Productivity is Lower
than Entrepreneurs Expectations, was 2.73. All three variables showed that the Thai logistics workforce
lacked the necessary logistics management skills and knowledge and the Thai entrepreneurs agreed that they
all contributed to higher operation costs and lower customer service. The skills, experience and knowledge of
logisticians are core competencies and increasingly important for logisticians in either management or at
educational institutes, therefore logistics knowledge and skills should be an important part of career path
development (Dischinger et al. 2006). Logistics skills are increasingly important for employment and career
development and the different skills requirements for different logistics activities have been concentrated
since specific logistics activities really need specific skills and knowledge (Kovacs et al. 2012). New
employees were not fully aware of specific requirements that were observed in the warehouse manager
recruitment process, such as testing the candidate on basic activities such as order picking (Murphy & Poist,
1993).
58
59
60
61
62
63
64
Criteria
Ranking
7,5
(Achievement orientation and ambition)
7,75
3
4
8
(Perseverance)
8,25
8,5
(Action orientation)
8,75
7
8
9
(Hardworking)
9,25
(Determination )
9
9,5
(Creativity)
Criteria
Ranking
7,25
(Creative problem-solving)
7,5
(Persuading)
7,75
(Negotiating)
4
5
(Proposing)
6
7
8,25
(Holistically managing business/projects/situations)
8,5
(Strategic thinking)
8
9
8,75
(Selling)
(Networking)
65
Criteria
Ranking
4,25
(Business plan)
(Financial Skills)
4,5
4,75
(Marketing Skills)
(Operational Skills)
5
5,25
5,5
(Legal Skills)
(Communication Skills)
5,75
6
(Leading Skills)
6,25
(Management Skills)
Based on the overall results of respondents' statements to the criteria of entrepreneurial learning models for
new entrepreneurs.Showing that the criteria used for selecting entrepreneurial consists three criteria:
motivation, entrepreneurial skills and business skills. Motivation is the highest criteria that must be
possessed by the prospective entrepreneurs because of the presence of a strong motivation will most likely
have a strong ability to do business. Entrepreneurial skills have value both of the criteria assessed.
Entrepreneurial skills are skills that should be possessed by every entrepreneur. This skill is mainly
associated with the skills to manage the business to be run. The third Criteria are business skills. These skills
are categorized into important criteri also to be possessed by the entrepreneur, which includes the ability to
manage the business internal and extern. Having a business skills course, be important as the existence of
businesses owned. The bigger the business, the higher the demand for manage to business
4.Conclution and Recommendation
Based on the results of the discussion of several conclusions can be drawn:
1.The success of entrepreneurial learning process is carried out by taking into account various factors,
namely:Iinstructo,course to be delivered,methode that used,techniques and tools that will be used and
the feedback is expected to learning process
2. The criteria used for the selection of prospective entrepreneurs consists of three criteria:
a. motivation
b. entrepreneurship skills
66
67
68
standard
errors
&
Coefficient
-0.00316
-0.02039
-0.18109 *
-0.00084
0.246268 *
0.001328 *
-1.96E09
0.415242
0.381506
6.44
4. Results
For the purpose of analysis two types of regression analysis was attempted. The first is least squares with
white heteroskedasticity consistent estimators. This was down with purpose of controlling for possible
problems posed by heteroskedasticity which is likely to occur in cross section data. The results show that R
squared and adjusted R squared are good and that average receivables, gross profit margin and age are
statistically important variables explaining behaviour of profit before tax to total assets, that is pre-tax ROA.
69
Coefficient
-0.00963
-0.00788
-0.19909 *
-0.00095
0.206834 *
0.001684 *
7.75E-10
Robust Statistics
0.271762
0.229749
67.74606
5.86
The results obtained are same as those of the White - adjusted least square procedure. Again we found that
average receivables, gross profit margin and age are the significant variables explaining the behaviour profits
(pbtta). Average inventory levels and average payables had no influence on firm performance. The negative
sign indicates that higher the average receivables lower is the pre tax ROA. Further Jarque-Bera test for
normality indicates that the errors are normally distributed indicating that our model does not suffer from
mis-specification errors. (JB at 5.8 is less than the chi square value). The main findings of the study are that
corporate financial performance of GCC manufacturing companies, as measured by pre-tax return on assets,
is strongly influenced by average accounts receivable levels after adjusting for the influence of control
variables. Inventory levels had no impact on financial performance. Similarly payables had no effect on
financial performance, although in theory higher levels of payables should lead to better profitability.
Average accounts receivables have a strongly significant negative influence on performance of GCC
manufacturing companies. Lower levels of average accounts receivable imply that the firm needs less
borrowed funds and that interest expenses are lower. This in turn directly results in better profits along with
less of assets, thus leading to better pre-tax return on assets. The results imply that a typical GCC
manufacturing company can improve profitability by reducing the level of accounts receivable. The paper
therefore concludes that better accounts receivable management will be beneficial from a performance point
of view.
References
Deloof M. (2003). Does Working Capital Management Affect Profitability of Belgian Firms? Journal of
Business. Finance and Accounting, 30, 573-587.
Heath. L.C. (1989). Is Working Capital Really Working?. Journal of Accountancy, August, 55-62.
70
71
Leadership Style Analysis of contingency and Demographics Map of Indonesia chose Jokowi
public interest, in the election of the President of the Republic of Indonesia Term of Office
2014 - 2019 (Case Study in Jakarta Special Capital Region)
Rr Dyah Eko Setyowati, A. Yani Antariksa, Sri Purwati
Sekolah Tinggi Ilmu Ekonomi Bisnis Indonesia
rrdyah@yahoo.com
Abstract: Contingency leadership style focuses attention on the match between the characteristics of the
leader's personal character, behavior and variable situation, assume that the situation is different, then the
contingency leadership style focuses attention on the broader aspects of the relationship between the
condition of a situational variable with a character or behavior and criteria performance leaders (Hoy and
Miskel, 1987). Sound as a quantitative measure of its election victory. Sounds a Millionaire is equal to a
laborer's voice. Given this reality, plus the implementation of majority voting system, can predict who is
elected as the President of the Republic of Indonesia are those who understand the character of the people.
Successful presidential candidate Tim and should know this. Factors that influence the character and
development of the population not only demographic factors. Yaukey (1990) says, demographic variables will
often be associated with a variable reciprocal nondemografi. One such relationship that gave birth to the
political demographic study of the population and the political aspects of the relationship. There are three
important political demographic variables. First, the number of inhabitants. Each area with a large number of
people would have a greater number of voters. Second, the structure or composition of the population. The
composition of the population can be observed in terms of gender, age group, economic, and education.
Jakarta population consisting Male 4.869.203, and 4.735.126 female residents. The study population was
2,859,894. (53.08%) of Jakarta residents who choose Jokowi on 9 July 2014 presidential election
(Commission, 2014). The study sample of 500 people, and kuantitaf qualitative data, primary and secondary
data. Processing data using multiple linear regression. From the data processing result that contingency
leadership styles and demographic maps significant effect on the interest of the people choose the President
of the Republic of Indonesia Jokowi period 2014-2019.
Keywords: Leadership Style Contingency, Demographic Maps, Election of the President of the Republic of
Indonesia
1. Introduction
Indonesia is are public consisting of more than 17,000 islands, which is based on the 1945 Constitution and
Pancasila. Indonesia is the number 4 in the world, which has a population of over 300 million people.The
presidential electionis conducted everyfiveyearsbydirect election. OnJuly 9, 2014, Indonesia heldthe election
of Presidentand VicePresident, for theperiod2014-2019.Prospectivejobcandidates forPresident
andVicePresidentatcooptationpresidentthis timethere weretwocouples, that PrabowoandHattaRajasa,
JokoWidodo(Jokowi) andJusufKalla(JK). The winner ofthe election of Presidentof the Republic ofIndonesiaon
July 9isthe pairJokoWidodo(Jokowi) andJusufKalla(JK). Official resultsJokowi-Jkvictoryannounced bythe
General Elections Commission(KPU) on July 22,2014.With53.08% of votesforpresident and
vicepresidentIr.H.JokoWidodoandH.JusufKalla.
And46.02%
of
the
vote
for
the
pairPrabowoSubiantoandHattaRajasa. Indonesian public recognizes.Ir. H. JokoWidodowho nicknamedJokowi,
ketkaelected governor ofJakarta. In the periodof GovernorJokoWidodo(2012-2014)or commonlyknown
asJokowi,
manyhandlingis
donebythe
city
governmentassessedto
fruition.
LeadershipPresidentJokowiduringbecomethe Governor of Jakarta, among others, thesuccessfulrelocationof
street
vendors
inTanah
Abang
marketcentersandotherstreet
vendorsinJakarta,
buildMRT,
buildreservoirsRiaRioas acommunityrecreation center, buildPluitdam, buildcheapflats, etc.
LeadershipJokowialwaysworkbased onpreviouslyestablished laws, among others, are:LawNo. 32 of2004
onRegional Government, JakartaRegional RegulationNo. 8 of2007 onPublic Order, and theGovernor
RegulationNo.221of 2009 onGuidelineslaw No.8 of 2008About thePublic Order. Efforts torelocatestreet
vendors inTanah Abang marketand elsewherebythe city governmentdone withsome changes inits
72
73
74
(X1)
75
76
From the results ofmultiple linearregressionanalysisof this study, it was also foundthe R-square value ofthe
regression equationis equal to: 0.382; This may be explainedthat the increaseoraddition ofone (1)value ofthe
dependent variablescoresintereschooseJokowi(Y) diakibatkaanbythe independent variablesof this
studyof0.382or38.2%. While the restof 61.8% caused by theotherindependent variablesoutside the
modelregression equationabove.
1.Test - F And Test - t
Proof of the data analysis of the study can be summarized and shown in Table 2:
Table 2: OutputTest-F
77
which
78
79
Influence Tourism Motives, the Tourism Component of the Interest of Foreign Tourists
Visiting the Beach in Indonesia, (2014)
RR Dyah Eko Setyowati, Sri Kartika Sari Antariksa
Stie Bisnis Indonesia
dyahantariksa@gmail.com
Abstract: This study discusses the factors that attract foreign visitors to visit the beaches in Indonesia.
Factors that affect the tourism in Indonesia. There are many reasons why they want to come to Indonesia,
namely internal factors and external factors. Indonesia has many islands that can be visited. But people
usually do tourism in Indonesia, on the island because we all know that Bali has a lot of very beautiful island.
Actually Indonesia not only has the island of Bali. There are 5 biggest islands in Indonesia, such as Sumatra,
Java, Kalimantan, Sulawesi, Papua. This study only discusses the factors in eastern Indonesia such as
Sulawesi, Nusa Tenggara, and Maluku Island. From the statistical data Number of foreign tourists coming to
Indonesia in July 2014 as many as 777 210 people. The survey was distributed in July 2014 with 500
respondents for tourism motives, but who has visited all the beaches of eastern Indonesia as many as 40
people so this sample to 40 respondents for tourism component. And the population in this study are the
tourists who have visited the coast to the tourist attractions in East Indonesia. The questionnaire was filled in
by foreign tourists. Methods ofdata processing wears multiple linear regression. Factors studied were the
motives of tourism and tourism components. Four motifs that attract visitors to visit tour places, (1) Physical
motivators (2) Cultural motivators, (3) Interpersonal motivators, (4) The status and prestige motivators.
Components that affect the tourism went to visit for a tour of the place is (1) Sightseeing and activities, (2)
accommodation, (3) facilities and other tourist services, (4) transportation, (5) other infrastructure, (6)
institutional elements. The results of the study did not affect the tourism motives of foreign tourists who tour
the coast of Indonesia. Factors that attract foreign tourists go to the beach resort is a component of the
tourism plan. Components such as a tourist attraction and activities, accommodation, other tourist facilities
and services, transportation, other infrastructure, institutional elements. The government should develop
these components so that foreign tourists can enjoy the beach resort in Indonesia.
Keywords: tourism motives, tourism component, interest of foreign tourists
1. Introduction
Indonesia is the largest archipelago in the world that has 17,508 islands. Indonesia between 6 degrees north
latitude and 11 degrees south latitude, and from 97 degrees to 141 degrees east longitude and is located
between two continents, Asia and Australia / Oceania. This strategic location has a profound effect on the
development of cultural, social, political, and economic. Scattered throughout Indonesia 3977 miles between
the Indian Ocean and the Pacific Ocean. When the waters between the islands be connected, Indonesia to 1.9
million square miles. Five major islands in Indonesia are: Sumatera with an area of 473,606 square km, Java
with an area of 132.107 square km, Kalimantan (the third largest island in the world) with an area of 539,460
square km, Sulawesi with 189.216 square km, and Papua with an area of 421,981 square kilometers.
(http://www.indo-journey.com/all-about-indonesia.php)
Indonesia has a tropical climate, and archipelagic geography, support the world's second highest level of
biodiversity (after Brazil), and its flora and fauna is a mixture of Asian and Australasian species. Once linked
to the Asian mainland, the islands of the Sunda Shelf (Sumatra, Java, Borneo, and Bali) have a wealth of Asian
fauna. Large species such as the tiger, rhinoceros, orangutan, elephant, and leopard, were once abundant as
far east as Bali, but numbers and distribution have dwindled drastically. Forests cover approximately 60% of
the country. In Sumatra and Kalimantan, these are predominantly of Asian species. However, forests are
smaller, and more densely populated Java, have largely been removed for human habitation and agriculture.
Sulawesi, Nusa Tenggara, and Maluku-that have long separated from the continental mainland-have
developed their own unique plants and animals. Papua is home to a unique fauna and flora closely related to
that of Australia, including over 600 bird species. In Indonesia, there are so many beautiful places and
upward, including the beach. Korem beach Papua, Pantai Kasih( Aceh), Bunaken ( Sulawesi), Raja Ampat
80
81
82
Cultural Motivators ( X2 )
Interpersonal Motivator ( X3 )
3. Methodology
Research Design: The research is a quantitative study using an explanatory design to prove the hypothesis.
The analysis model is a simple linear regression model. Tourists wear to components of qualitative research
data is obtained from direct interviews with respondents at the time of filling the questionnaire.
Population and Sample: The population in this study were tourists visiting the tourist attractions in
Indonesia during July 2014, the amount of study population 777 210 people, For the study, the study sample
size was 500 people (Sekaran, 2003). Respondents are the tourists who already knew and had visited the
beach the beach in Indonesia, so that respondents understand about the condition of coastal tourism in
Indonesia. This research attempts to analyze and to determine what factors are attracting foreign tourists to
visit the beaches in Indonesia.
Sampling Design: Sampling design will be used by researchers is probability sampling. The elements in the
population have some known chance probability of being selected as sample subjects. Probability sampling
design is used when the representativeness of the sample is important for broader generalization purposes.
The probability is based on the concept of random selection - a controlled procedure that assures that each
element of the population given a known non-zero chance of selection (William E, 1985). In this study,
researchers used simple random sampling to distribute the questionnaire and collect data. In confined or
better known as simple random sampling, every element in the population has a known and equal chance of
being selected as a subject. Thus, the authors randomly chose 500 visitors who visited Indonesia. Multiple
regression analysis is used by researchers to describe the effect of two or more independent variables on the
dependent variable. Multiple regression analysis is mathematically formulated as: Y = a + b1x1 + b2x2 + e
(Cooper and Schindler, 2006, p. 617)
4. Analysis and Interpretation
Test questionnaire through reliability testing of each of the variables used in this study by using the method
Gronbach Alpha method. This method is used to look at the consistency of each instrument used in each study
variable as outlined in the form of the questions in the questionnaire. Terms reliability testing with this
method is that if proven or known that the Grondbach (Alpha) of each variable is greater than the critical
value (r) product moment, then the variable is considered to be sufficient to say "reliable". Conversely, if the
value Grondbach (Alpha) of each variable is less than the critical value (r), then the variable is said to be
considered "unreliable" (Santoso, 2001: 316) Grobach of Reliability test results.
83
N of Items
3
N of Items
3
X2=Interpersonal motivators (Cultural Motive) grondbach Alpha value = 0.755> 0.6 means that reliable
Reliability Statistics
Cronbach's
Alpha
.735
N of Items
3
X3= Interpersonal Motivators (Cultural Motive) grondbach Alpha value = 0,621 > 0,6 means that reliable.
Reliability Statistics
Cronbach's
Alpha
.708
N of Items
3
X4= Interpersonal Motivators (Cultural Motive) grondbach Alpha value = 0,708 > 0,6 means that reliable
Reliability Statistics
Cronbach's
Alpha
.657
N of Items
3
Y= Interpersonal Motivators (Cultural Motive) grondbach Alpha value = 0,621 >0,6 means that reliable
Data Analysis: Further testing will be done with the research data sequential steps as follows
Table 1
Model Summaryb
Model
1
R
.649a
R Square
.421
Adjusted
R Square
.417
Std. Error of
the Estimate
1.83772
DurbinWatson
1.922
From the SPSS calculations, show that the value of R Square is 0.421. This suggests that a person's decision to
make the trip can be explained using existing independent variables. The residue, amounting to 57.9% (from
100 to 42.1%) can be explained by factors beyond investigated. In other words, the influence of the
independent variables the interest of foreign tourists visiting the beach in Indonesia amounted to 42.1%
person.
84
Model
1
(Constant)
Physical_Motivator
Cultural_Motivator
Interpersonal_Motivator
The_Status_and_
Prestige_Motivators
Unstandardized
Coefficients
B
Std. Error
4.965
2.642
.461
.049
.426
.048
.499
.051
.421
Standardized
Coefficients
Beta
.051
.323
.308
.335
t
1.879
9.404
8.922
9.690
Sig.
.061
.000
.000
.000
.286
8.324
.000
Collinearity Statistics
Tolerance
VIF
.990
.981
.979
1.010
1.020
1.021
.991
1.009
Regression
Residual
Total
Sum of
Squares
1216.473
1671.727
2888.200
df
4
495
499
Mean Square
304.118
3.377
F
90.050
Sig.
.000a
ANOVA test gives F number of 0.000 with a significance level (probability numbers) to 0,000. Because the
number of probability is 0.000 <0005, then the regression model is suitable for use in predicting the variable
Y.
In
other
words,
the
independent
variables
jointly
affect
the
variable
Y
From the table, it can be used to test the first hypothesis; with step do as following steps:
0.00 <0.05 or 5%. So we can say that X1 = Physical Motivator, X2 = Cultural Motive, X3 = Interpersonal
85
86
87
88
89
90
Leadership
Training Programs
- Credit Risk
Use of Technology
Entrepreneurship Orientation
Accounting Information
Source: Berg, 2010; Deakins, Logan & Steel, 2013; Dionne, 2003; Duong, 2005; Fatimah-Salwa, Azahari
& Tamkin, 2013; McClelland, 1975; Rauch, 2000; Virdi, 2005.
Leadership is referring to the small business owners they are the one who runs the business. The
appointment of the Risk Officer or Internal Auditor is rarely happens for small business due to the financial
constraints so the responsibilities lies under the owners themselves. By having particular documentation and
separate practices for risk management leads to effective financial risk management. When there are only
four or five employees in a firm, the owner usually has a much stronger impact on company policy, company
culture and the company's actions than in larger firms (Rauch, 2000). McClelland's (1975) early work on need
for achievement initiated many studies on characteristics of the entrepreneur. A high need for achievement
leads to prefer challenging tasks of moderate difficulty rather than routine or very difficult tasks, to take
personnel responsibility for one's performance, to seek feedback on performance, and to look for new and
better ways to improve one's performance. Many studies described the crucial role of training program,
contributes significantly and positively towards the success of entrepreneurs (Cheston and Kuhn, 2002; Jill et
al., 2007; Kuzilwa, 2005 as cited in Fatimah-Salwa, Azahari & Tamkin, 2013). Through the training program,
the entrepreneurs are expected to be well equipped in term of knowledge and skills necessary in effectively
managing their business and which subsequently result to their success. It starts from establishing goals
which refers to the threat of risk and its environment, identify, analyze the identified risks, assess or evaluate
the risks, manage the risks, monitor the risk and its environment regularly and lastly to continuously
communicate the risk strategies (Virdi, 2005; Dionne, 2003, Duong, 2005; Berg, 2010). It is viewed as a
medium for small business owners to learn and apply knowledge on how to act on risks besides find solution
to grow potential of risk to become an opportunity to grow. Moreover, training programs are where new set
of skills and information delivered from government or non-government bodies to help small business to
survive and commercialized their business (Andersen, 2013; Antonites, 2009 & Dinu, 2013).
Use of technology denotes between IT literate and IT illiterate. IT literate refers to the extent to which
technology benefits small business whereas IT illiterate refers to the small businesses that do not use the best
of technology. . According to Statistics of SMI (Small and Medium Scale Industries) Association of Malaysia,
only 30 per cent of the SMEs in Malaysia have a web presence and use IT on a daily basis (Hussin and Noor
2005 as cited in Alam, 2011). Due to the global reach of e-commerce, SME in the developed countries have
started adopting e-commerce in their businesses (Rao and Metts, 2003) but SMEs in Malaysian and many
other developing countries are still reluctant to use information technology or e-commerce in their day-today business operation. Also, there is evidence that many small businesses may not be keeping records as
well as they should and that those who keep records do so only to meet minimal reporting requirements. This
91
92
93
94
95
96
2. Literature Review
Job Satisfaction: The theory of two- factor or two -factor theory has been divided into two parts. One is
called the hygiene factors and motivator factors or second called 'hygiene and Motivational Factors. Herzberg
said both of these factors are not independent of each other. He stated that a high motivator factors will be
97
Stimulate Or Motivator
Advancement
Development
Responsibility
Award
Performance working group itself.
There are three things that need to be dealt with according to Herzberg's motivation in using:
1. Things that motivate employees is job accounting; sense of achievement, responsibility, advancement,
enjoy the job itself and the recognition of all.
2. Things that frustrate employees is especially a factor in employment, work rules, information, recreation,
reference department, rights, salary, and other benefits.
3. Employees will be disappointed when the opportunity for them to performing limited or restricted the
likelihood that they tend to be guilty of the offense more often.
The presence of the stimulant factors at work such as achievement, the work itself, responsibility,
improvement, growth and recognition will satisfy or motivate employees. However, if they do not, it does not
involve unhappy workers or to a member does not have the motivation to perform well in their jobs. On the
other hand the presence of health or hygiene factors such as supervision, job security and wages could lead to
employee discontent, but if this is not available, the dissatisfaction will occur. In short, the stimulus or
motivator factors can be identified as a factor that will lead to job satisfaction, while the health or hygiene
factor known as factor of discontent. Job satisfaction is often identified as the main reasons why employees
quit work of an organization, and working in other organizations. Thus, the ideas of allowing human Herzberg
satisfy their needs through employment and also help organizations achieve their mission. As a result, he
came out a trend in which the organization will strive to ensure that they work in a more satisfying and also
allow more rewards to employees
Organizational Commitment: Meyer, Allen and Smith (1993) formulated a definition of organizational
commitment as a psychological construct that is characteristic of the organization of labour relations with the
organization and has implications on the decision of individuals to pursue employment in the organization.
Based on the definition of an employee who has a commitment to the organization will be better able to
survive as part of the organization when compared with workers who did not have a commitment to
organization. Meyer and Allen (1984) have also defined organizational commitment into three components:
affective, continuous (continuance) and normative.
Affective Commitment: Affective component is similar to the psychological approach by Porter (1974) who
stressed the unity and the values shared by the employees in the organization. Identification refers to the
sense of unity (unity) of employees in the organization. Employees with a strong affective commitment
remain in the organization for their own needs and desires. The results of several studies Angle and Perry
(1983), affective commitment to the organization proved to correlate with age and working time . According
to the study Cherrington (1994) find between age relationship and commitment due because older workers,
more committed to the organization as well as older workers have or feel a positive experience with the
organization. Analysis of the age did not show the same effect, but studies (Gould, in Meyer & Allen, 1997)
shows that the relationship between job satisfaction and job distress is stronger perceived by younger
workers compared to older. It is possible there is a relationship between organizational commitments by
employees of different ages. According to Meyer and Allen (1997) related to affective organizational
commitment causes include individual characteristics, organizational characteristics, work experience, but
according to Meyer and Allen (1997), indicate that the strongest evidence found at the cause of the working
98
99
100
101
102
PPI Revenue
Strength (%)
PPI
(%)
(b a)
(b c)
National Park
447060 (1)
172800 (1)
39
45
Seafood Restaurant
211000 (2)
54000 (3)
26
14
170000 (3)
42000 (5)
25
11
Homestay
54
146880 (4)
61615 (2)
42
16
Lekor Piai
142000 (5)
52800 (4)
37
14
Total
60
1116940
383215
Average 34%
100
Share
103
104
105
106
107
108
109
110
111
112
113
The realm of economics explicitly distinguished between the informal sector with the informal economy (JJ
Thomas in Yustika, 2000: 190). For the context of the informal economy, there are at least four production
sectors where the informal sector is one of its parts. The informal sector is characterized as a small-scale
producers, using their own labor for the production of goods as well as many involved in business activities,
transport and service provision. Usually the output of the informal sector is sold as goods and services
between (intermediate goods and services) to other manufacturers or as final goods (final demand) directly
for consumption and thus already contained in the informal sector market. The most important thing to note,
that all goods and services produced in the informal sector are legal, although in general there are no rules in
the process of production and distribution. Studies conducted by Soetjipto (in Yustika, 2000: 194) on the
informal sector in the context of Indonesia produced characteristics of informal sector activities include
irregular pattern, both in terms of time, capital and acceptance, not touched by the rules or regulations set by
the government , capital equipment and supplies as well as the turnover is usually small and arranged on the
basis of the daily count. Generally do not have any other big businesses, generally done by and serve the lowincome segments of society, does not require special expertise and skills, each unit generally employ a little
effort and environmental relationships of family, acquaintances or from areas same, and do not know the
banking system, bookkeeping, credit and so forth
RelationshipInformal SectorandFormal Sector: Urbandualisticconditionoccurswith the development
ofanurban area.Dualisticconditioncan be seen fromthe emergence ofthe termformal sectorandthe
informalsector.
The
formal
sectorincludes
companiesthat
have
alegal
status,
recognitionandofficialpermission,
andgenerallybeskalagreat.
While
theinformal
sectorbusiness
activitiesgenerally simple, does not havea business license, generally lowincome levels, linkageswithother
businessesare verysmall, diversifiedits business, and thebusiness scaleis relativelysmall(Simanjuntak, 1989).
According toRamli(1992) that theinformal sectoras a providerother thanfield of jobalsowherethe informal
sectoristhe abilityto surviveinthe urban without drivenofgovernmentisbecause ofthe need for awide range of
productsandservices
producedby
the
informal
sectorini.Anyassumed
that
theinformalsectorformalneedcondition soexactly rightto say thatthe formal sectorandinformal likerelatedand
complementaryinurbaneconomic activities. One form of informal sector trade is so important is the street
vendors. Even so important and since has in the informal sector, informal term always identificated with the
type of work performed by vendors. On the other hand the activities of street vendors turned out to give a big
contribution in economic activity and well-being of people, especially in economically weak. In addition, the
activities of the informal sector is a characteristic that is independent of people's economy and concerns the
lives of many people. Consider the potential condition and proper handling practices and training activities of
street vendors should be based on the concept of environmentally friendly behavior and characteristics in
order to fill in the appropriate settings. Most of the street vendors and the surrounding urban region are not
natives (immigrants from outside the village or province), and is not a first choice as a livelihood.
3. Methodology
Berdasarkan tujuan yang ingin dicapai, tipe kajian ini merupakan penelitian deskriptif komparatif
(comparatif descriptive research), yang dilakukan dengan melakukan survey. penelitian survei adalah metode
penelitian deskriptif yaitu metode penelitian untuk membuat gambaran suatu kejadian. Metode survei
dilakukan bila data yang dicari sebenarnya sudah ada di lapangan atau obyek penelitiannya telah jelas. Data
yang digunakan terdiri dari data primer dan data sekunder, baik yang bersifat kualitatif maupun kuantitatif,.
Unit analisis dalam kajian ini adalah pada pedagang informal di Kota Depok. Penentuan lokasi sampel
dilakukan dengan sengaja (purposive), sedangkan penentuan unit analisis dilakukan dengan dilakunan secara
sensus kepada seluruh populasi pedang informal. Data yang dianalisis berdasarkan dimensi waktu dan ruang
pengumpulan merupakan data suatu waktu tertentu (cross sectional).
4.Result
Characteristics
Informal
Traders:
Based
on
theprimary
data
collectedas
many
as199peopleTradersinfromaltimeconsisting offive58informal tradersroad sectionToleIskandar, 67informal
tradersinJalan Juandaand74informal tradersinJalanSentosaDepok city, it is knowncharacteristics
114
115
116
117
118
Characteristics and Role of Women at Small Industries Manufacturing Leading in SumedangIndonesia and gender inequality are felt
Umi Zuraida, Ahmad Trias, Sri Herliana, Sri Hartati
Institut Technology Bandung (ITB), Indonesia
Winaya Mukti University
umi.zuraida@sbm-itb.ac.id
Abstract: The purpose of this study is to know the characteristics, role and gender inequality that was
received in a workplace and in the home by women who work at small industries manufacturing leading in
Sumedang district. Result of study can be used as a reference in realizing the government's policy of
increasing role of women's doubles at the small industries. Methodology : the research use descriptive
analysis. Sample are 60 woman who work at 15 main small industries that was taken by purposive. Primary
data with questionnaire accompanied by interviews and direct observation. The results : Women who
engaged in small industries in manufacturing commodity Sumedang district have characteristic : all within
the productive age with the highest number (76%) between 16 -30 years. Majority (60%) from primary
school and 90% earn below the minimum wages. 50% of respondents turn over all the proceeds to the
family, because motivation of their work is 53% to help their husbands in order to improve economy of
family. Contribution to the family economy are 60% contribute very small, because their contribution just
becomes part of the contribution of all other family members. Their role is mayority (60 %) at finishing and
packing. The respondent feel there are gender equality at the workplace, but there are still inequality in the
family even though they not feel that it is an inequality.
Keywords: Characteristic, Role of Women, Small Industries, Industries Manufacturing leading, gender
inequality
1. Introduction
Problem back ground: In Sumedang-Indonesia, The economy driven by Small scale Industriesl which
contributed 58.47 percent, including the Small Industries that were producing main Commodity who has
accounted for 80 percent of the contribution of the regional economy. Small industry and handicrafts which
mostly is a business legacy of more than 70% of workers in this industry are unpaid workers, including
women workers. This indicates that most of these businesses are family businesses and there are assisted by
family members as employees (BPS, 2003). The division of labor between men and women often causes
gender bias where men are culturally and structurally not obligated to work domestic area because it is the
responsibility of women. But when a woman entered the productive workforce area, it is considered
legitimate and reasonable even implicitly implied an obligation to help economic of family and must fixed
perform domestic work. This suggests that the dual role of women has been accepted by the society, but the
dual role of men have not. This study conducted to realize the government's policies in order to improve the
dual role of women, especially in small industry that has shown her strength in the face of economic
recession, especially in districts Sumedang, Gender bias resulted in a woman who works productively yet all
recognized as workers. Moreover, women who do work in the home (home industry), he can not separate
himself to the role of gender in the domestic sector. Gender bias also resulted in the participation of women is
at such a low position as a workers, employees, family workers or as unpaid family workers dependent on the
husband (Wijaya, 1994). This inequality must be looked and anticipated by the government, so the female
workers who participated in regional and national development are no longer in a marginal position, did not
understand his rights, not independent and treated less human.
Research objective: The purpose of this study was to know the characteristics, role and gender's inequalities
that was received in a workplace and in the home by women who work at small industries manufacturing
leading in Sumedang district of Indonesia.
119
120
121
Industries
1.
2.
3.
4.
5.
6.
122
123
124
125
126
Services Oriented Unified Theory Acceptance and Use of Technology (SOUTAUT): Tibenderana, Ogao,
Ikoja-Odongo, & Wokadala (2010) modify the UTAUT model and develop SOUTAUT to measure the level of
end-user acceptance and use of hybrid library services in eight Uganda Universities that have implemented
the technology in library services. SOUTAUT developed to replace some of the constructs in the UTAUT with
other constructs that are considered more suitable for library service environmental studies. In this context,
Effort expectancy" and "Voluntariness of Use" in UTAUT replaced by "Relevance" and "Awareness" to build
the SOUTAUT model. This was done to allow the study to determine the difference between the acceptance
and use of e-library services. Researchers considered "Relevance" and "Consciousness" more fit to build
digital library services. SOUTAUT model is shown in Figure 2 below.
Figure 2: SOUTAUT Model (Tibenderana et al., 2010)
127
A total of 116 students participated in this study and revealed that gender and nationality affects the
behavioural intention by social influences. Hence, the importance of social influence is more pronounced for
gender in the international category. Significant differences were found in the effects exerted on behaviour
intention by term performance between local and international groups. Overall, these results suggest that the
expected performance of the system shows that there is no distinction based on gender. Gender and
nationality will give a moderate impact on behaviour intention by expectations that efforts such as social
influence will be more pronounced for boys and girls in the country compared with their international
counterparts.
Figure 4: The influence of organizational culture on employee performance by Fakhar Shahzad et al.
(2013).
Organizational Culture: Every of human life in an organization where the organization is part and parcel of
it. According Gamble and Gibson (1999) which emphasis on the ability to achieve a common goal through the
effective and efficient communication between organizational units. Socialization process among members of
the organization occur in order to develop perceptions, values, and confidence in the organization (Kotter &
Heskett, 1992). Moreover, Stewart (2011) stated that the norms and values of the organizational culture can
128
Proposed Model: The research model is constructed based on a few model which are UTAUT (Venkatesh,
Morris, Davis, & Davis, 2003), SOUTAUT (Tibenderana et al., 2010), NUTAUT (Orji, 2010) and cultural model
(Fakhar Shahzad et al., 2013). However, the social influence that had been constructed by previous
researchers put that social influence as the extents to which individuals consider that others believe are
significant in order to use the new system (Venkatesh, 2003; Tibenderana, 2010 & Orji, 2010). Therefore, it is
not suitable to use in studies on the different school culture under MOE. This is because the organization
under the MOE has a unique ability to receive and executed instructions despite differences in many aspects,
especially the aspect of organizational culture. Sami Abdullah Al-bahussin and Wael Hassan El-Garaihy
(2013) had agreed with this issue and said that the organization's culture had also led to changes and
organizational innovation. By putting the organizational culture as a community then it does not stand in the
form of an individual. This is because in shaping culture, it must take a certain period of time as specified by
Lytle, Brett, Barsness, Tinsley, and Janssens (1995). Thus, culture can be represented as a series of ruts, or
inferred behavior pattern fundamental beliefs about past behavior and future. Therefore, this study can
provide a better understanding of information technology and system from cultural conflict perspectives
(Abd Latif Rahman, 2012) as proposed by UTAUT, where the sample was taken from a different cultural
environment.
129
130
131
132
133
Knowledge
Sharing
Technological
Innovation
Market
Orientation
134
135
136
137
138
139
140
141
142
143
(A)
The process of separation the right and wrong regularities from each other is very personal too and highly
depends on the personal experience and the individual point of view of the philosopher (researcher). In other
words, there is some Logical Separator (LS) hidden deep in the human mind that separates possible and
impossible regularities. Under common circumstances, a person rare understands the presence of such
thing in his/her mind. However, a person usually gives a certain answer on any question about possibility of a
given regularity. Each regularity that can go through the Logical Separator looks like a possible one from the
persons point of view. Otherwise, the regularity seems as impossible one. Strictly speaking, Logical
Separator depends on some philosophical system that the person uses. As a result, each philosophical system
establishes its own Logical Separator that protects the system from any regularity that seems impossible
in the logical frame of a given philosophical system. Such situation leads to the most dramatic result that
affects every attempt to use a new regularity or create a new theory.
The Logical Separator based on the generally accepted philosophical system
rejects a newly detected regularity as impossible one and denies any attempt of
further analysis of that regularity
(B)
Deep antagonism between an unusual regularity (or the fact) and a generally accepted philosophical
system leads to often rejection of such regularities and facts as impossible. They commonly make statement
that those regularities impossible by themselves. However, a real impossibility appears only in the human
mind as soon as regularity meets Logical Separator and becomes rejected by it. Such separation leads to one
important side-effect. A philosophical system accepts regularities and facts only for some extent that cannot
be expanded by the system. The most distant regularities and facts that keep compatibility with the
philosophical system establish the Comprehension Horizon (of that system and a person who shares the
system). In other words, anything inside the Comprehension Horizon seems logical, good and well. Anything
outside of the Comprehension Horizon seems illogical, strange and mysterious. As result, any additional
regularity inside the Comprehension Horizon leads to growth of a Philosophical System until appearance of a
regularity that stays beyond the Comprehension Horizon and becomes incompatible with the established
Philosophical System. Unfortunately, those regularities commonly treated as mysteries instead of close
physical and philosophical examination. In that case, they can be used as a basis of new and more
sophisticated Philosophical System. The best historical example of such transformation is a process of
changes the human imagination from a Philosophical System based on the idea of immovable flat Earth to a
new Philosophical System based on the idea of spherical Earth moving by its space trajectory (orbit) around
the Sun.
MOTION AND TRANSPOSITION: In case of Z-theory, the situation was the same. There were some mysteries
before publication of Z-Theory that were so strange that it looks impossible to recognize any regularity based
on those mysteries. All of them were facts and observations beyond the Comprehension Horizon of any
given theory existed at the time of observation. The best example of such phenomena was Boeing 727
incident. As a matter of fact. The ordinary aircraft was undetectable for the airport ground radar for 10
minutes during its descending by glide path. During that time, all attempts to reach the aircraft by radio have
failed. After 10 minutes the aircraft repapered again and have made successful landing. But as it was found
later all on-board timepieces indicated 10 minutes left from the indication of the airport clocks. Moreover,
nobody onboard had noticed anything strange or unusual. (Zade, 2012a)
144
Subsequently, there was not any investigator who was able to produce any suggestion about the incident. The
core problem was that. The Philosophical System accepted at the time of the incident based on the ideas of
endless existence of an object in Space and endless flow of so-called Time. Those core ideas and Logical
Separator based on them deny any possibility of existence of any object that jumps from one point of space
(and/or Time) to any other point of Space-Time under any circumstances. However, the incident becomes
analyzable in relationship with higher implementation of motion. From the one hand, the incident taken from
the Earth-bound observer looks like some jump through Time. From the other hand, the same incident,
taken from the point of view of a Sun-bound observer (who uses a Sun-bound reference frame) looks like
jump through Space. In a Sun-bound reference frame, the incident takes the following steps. There was
some point A where an aircraft coexisted with the Earth. Suddenly the aircraft became undetectable and
unreachable by any means (disappeared). The Earth continued its common orbital motion despite the
incident. 10 minutes later, the point above the Earth surface, where the incident had palace, reached the point
B (in the Sun-bound reference frame). Suddenly, the same aircraft reappeared at the point B. Therefore,
the incident looks like Space jump from a Sun-bound observers point of view (in the Sun-bound reference
frame).
Moreover, as soon as the same aircraft used the same points A and B, those points should be connected by
some sort of relationship (by a possible trajectory in common understanding) that caused a strange,
undetectable relocation of the aircraft. Later, that trajectory possessed a specific name of Z-Trajectory. The
noticeable aspect of the entire event was that. The aircraft had the same altitude before and after
disappearance in complete coincidence with the law of conservation. Subsequently, that law was extended to
the entire theory, and many phenomena were explained by application of the same law.
By definition given in the book,
Z-Trajectory is a special part of a moving objects trajectory that cannot be viewed or interacted with by the
IB (Independent Bystander) and always lies between two special points where the difference in the whole
energy of moving object equals zero. As a result, the whole energy of a system keeps constant before and after
an object uses Z-Trajectory. In other words Z-Trajectory is the image trajectory that connects two points of
space with equal value of full strength of conservative fields by means of a special number of points that have
no interaction with the surrounding conservative fields produced by independent bodies. (Zade, 2001, p.
240). Subsequently, motion of an object by Z-Trajectory became one of Z-Theorys cornerstones and
possessed a specific name of Transposition to be distinct from common way of Motion in common Space.
Later, that distinction was explained in details in the article Motion and Transposition in conservative fields
(Zade, 2012b)
A MATTER OF TIME: Everybody who tries to think about Time turns immediately to nonplus. The core problem
of that situation is this. There was not any suitable definition of Time before 2012. As a result, there was no
way to think about anything that has not any recognizable set of attributes. Therefore, as long as there is not
any definition of so-called Time there is not any way to make any conclusion about any category related to
Time. There is not also any possibility to make Right or Wrong decision about anything related with
Time. As a result, everything that the humankind knew about so-called Time falls under question. The
core illusion was that. According to the established Philosophical System, Time appears as some aspect of
physical reality of the Universe because it seems measurable by clocks. However, there is not any clock that
can make measurement of Time by itself. The easiest example for that is this. There is not any clock that can
make anticipated indication after interruption. Suppose, someone made interruption of operation of a clock.
In that case:
1.
2.
3.
145
(C)
(D)
(E)
(F)
(G)
What is Now?
Now is a point in the Universe from where an observer (object, body, etc.) makes
interaction with surrounding Universe. (Zade, 2012d)
(H)
The clear physical definition mentioned in the citation given above created the next cornerstone of Z-Theory.
THE MATTER OF NAVIGATION: Many phenomena mentioned above create a significant problem of navigation for
every man-made vehicle involved in motion on Z-Trajectory. The result of Transposition usually appears as
sudden mysterious relocation of a vehicle from one location to another one. The next example of
Transposition comes from the experience of Bruce Gernon, who was involved in Transposition with an
aircraft more than once. Later, Mr. Gernon made a significant effort to explain those incidents and proposed
his point of view on that matter in the book The Fog. (Gernon, 2005). Strictly speaking, Mr. Gernon
described some phenomenon that caused mysterious relocation of his A-36 aircraft. According to
information from Bruce his aircraft had air-journey for 250 miles in 47 minutes flying from Andros Island
Town Airport to West Palm Beach Airport and had successful landing with 9 extra gallons of unspent fuel.
(Zade, 2012a). Moreover, Mr. Gernon described some unusual optical phenomenon he experienced during the
flight.
In that case, A-36 used some Z-Trajectory as well as Boeing 727. The difference was that. In the second case
the planet didnt move during relocation of the aircraft in comparison with the aircraft relocation. As a result,
A-36 appeared at the end point of Z-Trajectory almost immediately after disappearance from the first point
of Z-Trajectory. The duration of relocation by Z-Trajectory for A-36 and Boeing 727 was the same and equal
almost to zero in both cases. The difference appeared only for Earth-bound observers. They did not notice
any difference in indications of onboard clock and Earth-bound clocks in the second case. In other words,
duration of A-36 Transposition and Motion of the planet during Transposition were the same. As a result,
both sets of clocks onboard and Erath-bound count the same duration of their internal recurrent processes.
Hence, indications of those clocks were the same before and after the Transposition in that case. The same
aspect of operation of a clock (or a watch) caused the age-old problem for the human mind. Any couple of
timepiece count only oscillations of their own recurrent physical process and nothing more. Durations of
those processes are the same ever for any number of Earth-bound timepieces. As a result, any numbers of
previously synchronized Earth-bound timepieces show the same indications as long as they keep identical
trajectories in the common space.
146
(7)
147
(I)
148
Forecasting returns on a stock market using Artificial Neural Networks and GARCH family
models: Evidence of stock market S&P500
Nadhem Selmi*, Samira Chaabene, Nejib Hachicha
Faculty of Economics and Management of Sfax Tunisia
nadhem.selmi@yahoo.fr
Abstract: In the area of financial stock market forecasting, many studies focused on application of Artificial
Neural Networks (ANNs). Due to its high rate of uncertainty and volatility, the stock markets returns
forecasting by conventional methods became a difficult task. The ANNs is a relatively new and have been
reported as good methods to predict financial stock market levels and can model flexible linear or non-linear
relationship among variables. The aim of the study is to employ an ANN models to estimate and predict the
dynamic volatility of the daily of S&P500 market returns. Results of ANN models will be compared with time
series model using GARCH family models. The use of the novel model for conditional stock markets returns
volatility can handle the vast amount of nonlinear data, simulate their relationship and give a moderate
solution for the hard problem. The forecasts of stock index returns in the paper will be evaluated and
compared, considering the MSE, RMSE and MAE forecasts statistic.
Keywords: Artificial neural networks, stock market forecasting, non-linear modeling, GARCH, Forecasts.
1. Introduction
Stock market prediction (or forecasting) is one of the hottest fields of research lately due to its commercial
applications owing to the high stakes and the kinds of attractive benefits that it has to offer (Majhi et al,
2007). In addition it is regarded as a challenging task of financial time-series prediction and one of the
important issues in finance. The stock market, which has been investigated by various researches, is
essentially non-linear, dynamic, nonparametric, and is rather complicated environment (Tan et al, 2005).
Miao et al (2007) and Wang (2002) presented that stock markets movements are affected by many factors
such as political events, firms policies, general economic conditions, commodity price index, bank rate, bank
exchange rate, investors expectations, institutional investors choices, movements of other stock market,
psychology of investors, etc. The explanation and the understanding adequate of the stock markets return
volatility constitutes a central to the study of finance. Investigating the data generating process in stock
returns, recent research has suggested that the factors affecting the asset pricing behavior of investors are
described by nonlinear relationships with expected returns. The predicate of stock markets returns have
exhibited a strong nonlinearity induced from an asymmetric dynamic process (Nam et al. 2002; Nam and Kim,
2003)1. Forecasting of stock market returns in todays volatile markets became a difficult task and represents
a major challenge for traditional time-series predicting. Therefore, predicting finance and economics
movements is quite difficult. In the literature there are a number of methods applied to describe the behavior
of time series and to accomplish this task.
In the last decade, in addition to the advert of the advancement in computer technology many studies have
been included soft computing techniques. ANNs2 have been popularly applied to every area of disciplines
financial type problems such as stock markets index prediction, exchange rate prediction, bankruptcy
prediction etc. This technique of soft computing can provide better forecasting ability on stock market than
other models, especially in nonlinear questions (Phua et al. 2003; OConnor and Maddem, 2006; Kunhuang
and Yu, 2006; Zhu et al. 2007). ANN models have been proposed as a promising alternative approach to timeseries forecasting. Many studies have showed that ANN can be a very useful tool for time series modeling and
forecasting (Zhang et al, 1998). However this novel model of soft computing has the capability to identify the
underlying functional relationships in the data and flexibly generate the forecasting models regardless of the
complexity of system. There have been many studies using ANNs in stock market prediction. These studies
They identify nonlinearities induced from an asymmetric reverting property of weekly and monthly returns
for U.S and Pacific Basin equity markets.
2 ANN: artificial neural networks.
1
149
t ht .vt
(2.2)
(2.3)
i 1
(2.4)
This model is also recognized as the linear ARCH (q) model. With financial data it captures the information
from t . Bollerslev (1986) introduce an alternative model, the GARCH (p,q) model. In his model, ht was
articulated as following:
Where
i 1
i 1
(2.5)
The ARCH-M model was introduced by Engle, Lilien and Robins (1987). While many theories in finance
engage an explicit trade off among the risk and the expected return, this model is preferably suited to
handling such questions in a time series background where the conditional variance may be time-varying.
The model is expressed as follows:
yt yt 1 ht t
(2.6)
People found the applications of the GARCH model limited because the conditional variance is simply linked
to past conditional variances. Then the EGARCH model was modified by Nelson (1991). In this class of ARCH
models, the volatility depended not only on the degree of the past surprises in returns but also on their
corresponding signs. EGARCH process is indicated as follows:
ln ht 0 j ln ht j i g vt i
p
j 1
i 1
(2.7)
Such as Chang et al (2004); Hamid and Iqbal (2004); Liu et al (2003- 2006), Chen et al (2006), Zhu et al
(2007).
4 GARCH: Generalized Autoregressive Conditional Heteroscedasticity.
3
150
g vt i vt
t
ht
(2.8)
ht
To sum up, we can find that the ARCH family models focus on structure the different models to dispose the
variance t in order to acquire more information from the time series data.
The Artificial neural networks model (ANNs): In this section, we brief describe the functional from ANNs 5,
which are considered best for predicting stock market time series. Artificial neural networks, as
computational intelligence systems, are nonlinear methods, taking an input vector X and produces output
vector Y. The relationship between X and Y is determined by the network architecture. They do not require
specific assumptions about the underlying model and has been used widely for approximation functions and
forecasting (Hornik et al., 1989; White, 1989). For this reason, ANNs are called universal approximates that
can approximate a large class of functions with a high degree of accuracy (Chen, Leung and Hazem, 2003;
Zhang and Qi, 2005). Newly, ANNs have been successfully applied to predict important financial markets, like
for example indexes, S&P500 and others (Chen, 1994; Huarng and Yu, 2006; Huang, Lai, Nakamori, Wang and
Yu, 2007; Yu and Huarng, 2008) and others applications (Jain and Kumar, 2007; Majhi et al, 2009). The
greatest advantage of an ANNs over other modeling techniques is their capability to model complex nonlinear relationships in the observations with a high degree of accuracy (Chen et al, 2003; Wong et al, 2000).
Among types of neural networks that are endowed to this property, they are the multilayer perception (MLP),
radial basis function (RBF) networks and recurrent networks, which are widely used in time-series
forecasting, because they are capable of resolving a wide of problems (Sarle, 2002). But the multi-layer feed
forward model is the most popular neural networks used in forecasting problems.
A multilayer perceptron neural network model, developing by the back-propagation algorithm, is grouped in
layers and a set of layers forms a network. There are three types of layers: inputs, representing the external
data into the artificial neural networks, output, which relays the information out and hidden layers that,
appears between the input and output. For input layers, it receives the inputs and directs them to the hidden
layers. For time series forecasting and modeling, one hidden layer feed forward network is the most widely
used (Zhang and Qi, 2005). The MLP neural network is illustrated in figure1. This figure shows an example of
a simple feed-forward network for time series analysis with one hidden layer. The input layer has five nodes,
and the hidden layer has three. The input nodes are connected forward to each and every node in the hidden
layer, and these hidden nodes are connected to the single node in the output layer. We call the network a 5-31 feed-forward network.
Input layers
Hidden layer
output layer
The first artificial neuron was proposed in 1943 by the neurophysiologist Warren McCulloch and in the 1990s the
use of neural networks for identifying or predicting time series became widespread.
151
1+exp
(x)
1exp
(2x)
(2.11)
1+exp
(2x)
In fact, the ANN model of equation (2.9), performs a nonlinear functional mapping from the past observations
to the future value yt , i.e.
yt = f yt1 , yt2 , , ytp , + t ,
(2.12)
Where, f (.) function is determined by the network studies and connection weights. is a vector of all
parameters. Getting up and development with Neural networks involves several steps: (Gradijevic and Yong,
2000).
1. Specifying the structure of the neural network or selection of the input and output variables, number of
hidden layers, and number of hidden neurons, training algorithm, activation function and initial weights.
2. Division of the data set into three sets, network training, validation and testing sets. Sixty percent of the
data set using to training, thirty percent was used for validation and the rest to test.
3. Estimation of the predict output. The input values of the last 10 percent of the observation were then
used by the trained to generate output, i.e. stock market returns forecasting in this study.
4. Evaluation of the forecast performance of the ANN, and comparison of it to other models using RMSE or
others measures.
5. Steps 1 to 4 were repeated until the error goal was not reached.
Over the last decade, neural networks represent a recent approach to time series forecasting that are a newly
proliferating technique in desktop quantitative analysis. They have been found to be a viable contender to
various traditional time series models (Chen, Yang and Dong, 2005; Jain and Kumar, 2007).
Forecast Evaluation criterions: Three criterions will be used to evaluate the model forecasting aptitude.
The first is the root mean squared error (RMSE). The formula for RMSE is:
n
RMSE
y
t 1
yt
(2.13)
MAPE
1 n yt yt
100
n t 1 yt
(2.14)
The third is the Theil disparity Coefficient (Theil IC). Theil IC value is always among 0 and 1, and a smaller
value designates the error among the predicted value and the actual value is smaller. Theil IC is given by:
152
y
t 1
Theil IC=
yt
n
n
y
t 1
2
t
(2.15)
y
t 1
2
t
Empirical analysis: The daily of S&P500 futures price data are obtained from Reuters Information System.
The data span the period from 8 July 1996 to 26 May 2006 (2488 observations). We take data from 8 July
1996 to 25 May 2005 as in-sample data sets ( 2236 observations), and we take the data from 27 May 2005 to
26 May 2006 as out-of-sample data sets ( 252 observations) which are used to evaluate the performance of
the predictions
Figure2. Plots logarithm returns data series of S&P500 futures price
.
The main equation is regression equation and Akaikes Information Criterion (AIC) rule is used to determine
the lag lengths. Finally, we select 3 lags after testing many times. The residual equation we selected are ARCH
(1), GARCH (1,1), ARCH (1)-M, GARCH (1,1)-M, EGARCH(1,1) for Bollerslevs (1988) idea, in most applications
p q 1 is found to suffice. On the other hand, the ANN model uses trial and error to determine the
network architecture of (12) by minimizing the forecasting error. First, we process the forecast using the
ARCH family models. Table 1 gives the forecasting result. According to the forecast evaluating criteria, we can
find that the ARCH(1)-M performs best.
Table 1: Forecast Result of ARCH Family Models
RMSE
Rank
MAPE
ARIMA
7.214
3
1.342
ARCH(1)
7.211
1
1.348
GARCH(1,1)
7.213
2
1.346
EGARCH(1,1)
7.216
5
1.340
ARCH(1)-M
7.215
4
1.343
GARCH(1,1)-M
7.217
6
1.345
ANN( )
6.38
4
1.26
Rank
2
6
5
1
3
4
4
Theil IC
0.0072
0.0074
0.0073
0.0075
0.0077
0.0079
0.0003
Rank
1
3
2
4
5
6
5
Therefore, we form the hybrid model using the best ARCH(1)-M model combined with the ANN model. Table
2 gives the forecasting results. Simultaneously, we also process the forecast using the ARIMA-ANN hybrid
model. According to the forecast evaluation criteria, we can find that the ARCH (1)-M-ANN model performs
better than the ARIMA-ANN model, and also performs the best in all the given models.
153
Rank
1
2
Theil IC
0.0064
0.0065
Rank
1
2
According to the NMSE, MAPE, Thiel IC evaluation criterion, we can discover that the ARCH(1)-M-ANN hybrid
model is greater to ARIMA, ARCH(1), GARCH(1,1), EGARCH(1,1), ARCH(1)-M, GARCH(1,1)-M and ARIMA-ANN
models for S&P500 futures price forecasting.
4. Conclusion
Commodity index futures represent a useful investment vehicle for speculators and hedgers in stock markets.
In this study, we suggest a hybrid model combining ARCH-M model and ANN model to predict S&P500
futures price. In terms of the empirical results, we find that the model combining ARCH-M and ANN models
performs the best on the selected criteria. In this paper, the auto-regressive integrated moving average
models are applied to propose a new hybrid method for improving the performance of the artificial neural
networks to time series forecasting. In our proposed model, based on the Box-Jenkins methodology in linear
modeling, a time series is considered as nonlinear function of several past observations and random errors.
Therefore, in the first stage, an auto-regressive integrated moving average model is used in order to generate
the necessary data, and then a neural network is used to determine a model in order to capture the
underlying data generating process and predict the future, using preprocessed data. Empirical results with
three well-known real data sets indicate that the proposed model can be an effective way in order to yield
more accurate model than traditional artificial neural networks. Thus, it can be used as an appropriate
alternative for artificial neural networks, especially when higher forecasting accuracy is needed.
References
Ahmadi, H., (1990) Testability of the arbitrage pricing theory by neural networks, Proceedings of the
International Conference on Neural Networks, 385- 393.
Bollerslev, T., (1986), Generalized Autoregressive Conditional Heteroskedasticity, Journal of Econometrics,
31, 307-327.
Bollerslev, T., (1988), On the correlation structure for the generalized autoregressive conditional
heteroskedastic process, Journal of Time Series Analysis, 9, 121-131.
Chen, C. H., (1994), Neural networks for financial market prediction, Proceedings of the IEEE International
Conference on Neural Networks, 2, 1199-1202.
Chen, A., Leung, M. T., & Hazem, D., (2003), Application of neural networks to an emerging financial market:
Forecasting and trading the Taiwan Stock Index, Computers and Operations Research, 30, 901-923.
Choi, H., Lee, E., and Lee, C., (2006) Neural Network Deinterlacing Using Multiple Fields, Lecture Notes in
Control and Information Sciences, 345, 970-975.
Choi, H., Lee, C., and Rhee, M. W., (1995) Trading S&P 500 stock index futures using a neural network, The
Third Annual International Conference on Artificial Intelligence Applications on Wall Street, 63-72.
Engle, R.F., (1982), Autoregressive conditional heteroskedasticity with estimates of the variance of U.K.
inflation, Econometrica, 50, 987-1008.
Engle, R. F., Lilien, D. M., and Robins, R.P., (1987), Estimating Time Varying Risk Premia in the Term Structure:
The Arch-M Model, Econometrica, 55, 391-407.
Gradojevic, N., Yang, J., (2000), The Application of Artificial Neural Networks to Exchange Rate Forecasting:
The Role of Market Microstructure variables, Working Paper n 2000-23, Bank of Canada.
Hamid, S. A., and Iqbal, Z., (2004), Using neural networks for forecasting volatility of S&P 500 Index futures
prices, Journal of Business Research, 57(10), 1116-1125.
Hornick, K., Stinchcombe, M., and White, H., (1989) Multilayer feedforward networks are universal
approximators, Neural Networks, 2, 359-366.
Huang, W., Lai, K. K., Nakamori, Y., Wang, S. Y., and Yu, L., (2007), Neural networks in finance and economics
forecasting, International Journal of Information Technology and Decision Making, 6, 113-140.
Huarng, K., & Yu, T. H., (2006), The application of neural networks to forecast fuzzy time series, Physica A,
363, 481-491.
154
155
The Carry Trade Returns and Decomposed Foreign Exchange Market Volatilities
Wenna Lu, Woon Wong
Cardiff Business School, University of Cardiff
wongwk3@cardiff.ac.uk
Abstract: Menkhoff et al. (2011) find that the cross sectional excess returns from the foreign exchange (FX)
market can be explained by the FX volatility risk very well, the theoretical support of this finding is Chen
(2003) version of Intertemporal Capital Asset Pricing (I-CAPM) model, which argues that risk-averse
investors also want to directly hedge against changes in future market volatility and therefore market
volatility risks are negatively priced. In this paper, we apply Menkhoff et al. (2011) model to a longer period
and a new sample. We find that their findings are consistent, that the FX volatility risk is able to explain the
cross sectional excess returns from the foreign exchange market under the theoretical support from Chen
(2003). Further, as inspired by Adrian and Rosenberg (2008), we decompose FX volatility into two
components: a persistent component and a less persistent component and we find that both components are
negatively and significantly priced when explaining the excess returns from the carry trade. This implies that
investors pay for insurance against increases in FX market volatility, even if those increases have little
persistence.
Keywords: The carry trade, Volatility risk, Volatility Decomposition, Component-GARCH model, Forward
Premium Puzzle
1. Introduction
This paper studies the cross-sectional pricing of volatility risk in the carry trade returns. The carry trade
refers to a trading strategy in foreign exchange (FX) market that borrows in currencies with low interest
rates and lends in currencies with high interest rates. The carry trade is a profitable investment strategy in
the sense that although it has made substantial losses during the recent global financial crisis, its losses are
relatively small when compared to the historic returns from the carry trade. The question is, given the very
liquid foreign exchange market and the existence of international currency speculations, why the carry trade
has been profitable for such a long time. The risk-based explanation believes that by borrowing low interest
rate currencies and investing in high interest rate currencies, investors load up on certain risks, the excess
returns from the carry trade are simply compensations to investors for bearing those risks. In this paper, we
follow the logic of risk-based explanation to detect these certain risks. Chen (2003) version of
Intertemporal Capital Asset Pricing (I-CAPM) model argues that risk-averse investors also want to directly
hedge against changes in future market volatility. In his model, expected returns for asset depend on risk
from the market return, changes in forecasts of future market returns, and changes in forecasts of future
market volatilities.
Following this theoretical model 1) Menkhoff, Sarno et al. (2011) build up the stochastic discount factor
which is linear in (1) the FX market return and (2) the innovation in FX market volatility. They adopt the DOL
factor of Lustig, Roussanov et al. (2011) as a measure of the FX market return and create the VOL factor,
which is similar to the measure of realized FX volatility, as a measure of the FX market volatility. Their proxy
for global foreign exchange volatility risk captures more than 90% of the cross-sectional excess returns in five
carry trade portfolios. We extend their model to a longer time period and one new sample, and we find the
results are robust. The inspiration of using Chen (2003) version of I-CAPM model to explain the excess return
for the carry trade is from the literature of stock market, among those literature, the first successful
implication is Ang et al. (2006) in which they use the aggregate market return and aggregate market volatility
to explain the cross-sectional excess return from the sorted portfolios in the stock market. The proxy of
aggregate market volatility they use is the implied volatility- the VIX data. Inspired by this, Menkhoff et al.
(2011) apply this model in currency market and find to be very successful as well. Rather than implied
volatility, they use realized volatility formed by aggregate the average daily absolute changes in exchange rate
across currencies into monthly observation. Given by their success of applying models in stock market to
currency market, it is very tempting and interesting to apply other successful models in stock market to
currency market.
156
+1 =
+1
( ,+1 )3 1
The pricing kernel described by equation above depends only on the consumption growth rate and the
aggregate market return. This specification includes Epstein and Zins (1989) recursive utility function
representative agent as a special case. The link between consumption and the changes in the investment
opportunity set is provided by substituting out the consumption growth rate using the aggregate budget
constraint. Following Campbell, Chen (2003) shows that a log-linear approximation of the aggregate budget
constraint gives three factors. The first factor is the market return; the second is the change in the forecasts of
future market returns. The third factor is the change in the exponentially weighted forecasts of future market
variances. The intuition behind this substitution is that an increased level of consumption today must be
financed by a high current market return, an increased forecast of future market returns or a lower
expectation of future market volatility. Chen shows that risk-averse investors also want to directly hedge
against changes in future market volatility. For an investor more risk averse than log utility, Chen shows that
an asset that has a positive covariance between its return and a variable that positively forecasts future
157
mt 1 1 b1 RX me ,t 1 b2 Vt 1
where RX m ,t 1 is the log market excess return and Vt 1 denotes volatility innovations.
e
Ang, Hodrick et al. (2006) employ changes in the VIX index to proxy for volatility risk, considered as a nontraded risk factor. They find that market volatility is priced in the cross-section of US stock returns and that
shocks with a higher sensitivity to volatility risk do earn lower returns. Bandi, Moise et al. (2008) consider
not only volatility, but also liquidity as a further pricing factor. They find that both risk factors are useful for
understanding the pricing of US stocks, but that volatility dominates liquidity when they are considered
jointly. Da and Schaumburg (2009) price several asset classes with a pricing kernel that is linear in the stock
market return and volatility innovations. Inspired by this literature, Menkhoff, Sarno et al. (2011) show that a
similar approach is helpful to understand the cross-section of foreign change risk premium as well. They
follow Ang et al.(2006) and assume that the stochastic discount factor is linear in two pricing factors: (1) an
aggregate FX market return, and (2) aggregate FX market volatility innovations. They follow Lustig,
Roussanov et al. (2011) and discuss returns of interest rate sorted currency portfolios rather than single
currency. They use a straightforward measure to proxy for global FX volatility and their proxy has similarities
to measures of realized volatility. (Andersen, Bollerslev et al. 2001). For the empirical analysis, they focus on
volatility innovations as non-traded risk factor. They find this factor captures more than 90% of the crosssectional sensitivity among portfolios and also they find that their volatility risk proxy outperforms different
measures of the market liquidity risk.
Menkhoff, Sarno et al. (2011)s model is very successful in explaining the cross-sectional returns from the
carry trade portfolios. So in this paper, we test if Chen (2003)s model is (1) robust across data periods and
samples (2) independent of various proxies of volatility. We start by applying the same asset pricing test on a
longer data period and one more sample. The results are found to be robust. Further and more importantly,
we test if the model works for different measure of FX volatility risk. Instead of using only the realized
volatility measure, we use the conditional volatility measure from a GARCH model as inspired from stock
market literature. In one step further, we follow Adrian and Rosenberg (2008) to decompose volatility into
long run and short run components and we find both components are negatively and significantly priced,
which indicates investors are willing to pay insurance against volatility risk, even if the risk has little
persistence. The idea that market volatility is subject to shocks at different frequencies is from Engle and Lee
(1999) and thus they develop the Component-GARCH (CGARCH) model which decomposes volatility into
permanent and transitory components. The CGARCH model has been widely used recently in both economics
and finance because separating the permanent and transitory risk premium could help us to understand the
source of uncertainty, and investment decisions heavily depend on whether this uncertainty is permanent or
transitory The CGARCH model has been widely used recently in both economics and finance.
Adrian and Rosenberg (2008) adapt their model into log-normal models of volatility (as suggested by Nelson
(1991)) and decompose market volatility into short- and long-run components. They find that prices of risk
are negative and significant for both volatility components implies that investors pay for insurance against
increases in volatility, even if those increases have little persistence. They also find that their benchmark
asset pricing model with the market return and the two volatility components as cross-sectional pricing
factor outperforms the model used by Ang et al (2006). Furthermore, they relate these two factors to a
measure of the tightness of financial constraints and to the business cycle to interpret the economics of shortand long-run volatility. Apart from Adrian and Rosenberg (2008), there is a large body of literature providing
evidence that the CGARCH model works better than the standard GARCH models in explaining stock market
volatility. Apart from its success in equity market, literature has shown that the CGARCH model is also a
superior volatility model for exchange rates, as it can distinguish the permanent and transitory volatility
components to describe volatility dynamics better than other GARCH models. (Glosten, Jagannathan et al.
1993; Black and McMillan 2004; Byrne and Davis 2005; Ghysels, Santa-Clara et al. 2005; Pramor and Tamirisa
2006; Simon and Amalia 2011). Black and McMillan (2004) find evidence of short-run and long-run
158
159
P4
P5
P6
DOL
HML
1.73
-1.40
3.13
3.53
-0.57
4.10
9.90
3.78
6.12
1.94
-0.32
2.26
12.89
5.27
7.62
1.43
-1.38
2.81
3.04
-0.38
3.42
7.14
0.95
6.19
1.53
-1.14
2.67
9.47
2.46
7.01
This table reports the time-series average of the average forward discount (annualized and in percentage
points), the average rate of depreciation (annualized and in percentage points) and the average log returns
(annualized and in percentage points). Portfolio 1 contains currencies with the lowest forward discount.
Portfolio 6 contains currencies with the highest interest rates. DOL is the equally-weighted average of all six
portfolio and HML is the portfolio constructed by short portfolio1 and long portfolio 6. Table 2 reports mean
(annualized), standard deviations (annualized) and skewness of currency portfolios sorted monthly on time
t-1 forward discounts. Sharp Ratio defined as average return per unit of standard deviation is also reported
and denoted as SR in the table. Portfolio 1 contains the 20% of all currencies with the lowest forward
discounts whereas Portfolio 6 contains currencies with highest forward discounts. All returns are excess
returns in USD. DOL denotes the average return of the six currency portfolios and HML denotes a long-short
portfolio that is long in portfolio 6 and short in Portfolio 1. We report both log returns and discrete returns
with and without transaction cost adjustments8. The discrete returns are used in asset pricing tests. For both
samples when moving from portfolio 1 to portfolio 6 and the HML portfolio, the average returns
monotonically increase and the skewness are almost monotonically increasing in absolute terms. This is
consistent with Brunnermeier, Nagel et al.(2009), they argue that the returns from high interest rate
currencies are negatively skewed which suggests that they are subject to crash risk. There is no clear pattern,
however, for the standard deviation.
Figure 1 plots cumulative log returns for the carry trade portfolio (HML) for both samples. Shaded areas
correspond to NBER recessions. Interestingly, carry trades in the OECD countries were more profitable in the
80s and 90s; only in the last 10 years did the inclusion of emerging markets currencies improve returns to
the carry trade. Also, the two recessions in the early 1990s and 2000s did not have any significant influence
on returns. It is only in the last recession that also saw a massive financial crisis that carry trade returns
show some sensitivity to macroeconomic conditions. Most of the major spikes in carry trade returns seem
rather unrelated to the U.S. business cycle. This coincides with Burnside et al (2011) which finds that most
standard business cycle risk factors are unable to account for returns to carry trades.
Transaction costs are deducted. The reason and the way of deducting transaction cost are shown in details in Section 7.
160
30
25
20
15
10
5
0
-5
84
86
88
90
92
94
96
98
00
02
04
06
08
10
Year
all countries
oece countries
Table 2 Descriptive Statistics: The table reports mean returns (annualized), standard deviations
(annualized) and skewness of currency portfolios sorted monthly on time t-1 forward discounts. We also
report Sharp Ratios (SR). Portfolio 1 contains the 20% of all currencies with the lowest forward discounts
whereas Portfolio 6 contains currencies with highest forward discounts. All returns are excess returns in USD.
DOL denotes the average return of the five currency portfolios and HML denotes a long-short portfolio that is
long in portfolio6 and short in Portfolio 1. We report both log returns and excess returns for both samples:
48-all-country sample and 29-OECD-country sample. Returns are reported both without transaction costs
adjustment (without b-a) and with transaction costs adjustment (with b-a). The transaction cost adjustments
are computed according to Table ?. The time period is spanned from 01/11/1983 to 30/09/2011.
48-all country sample
Log return (without b-a)
Portfoli 1
2
3
4
o
Mean(
2.6
3.1
%)
1.5
0.9 1
3
0
3
Std.
8.4
8.4 8.2
8.7
Dev.
8
6
5
0
Skewne 0.1
ss
2
0.2 0.0
0.4
4
9
3
SR
0.3
0.3
0.1
0.1 2
6
8
1
Log return (with b-a)
Portfoli 1
2
3
4
o
Mean
0.2
1.3 2.1
2.5
8
1
9
5
Std.
6.5
8.4 8.2
8.6
Dev.
4
5
4
9
Skewne
ss
0.2
0.2 0.0
0.4
5
4
9
5
SR
0.0
0.1 0.2
0.2
4
6
7
9
4.1
0
6.12
8.6
3
0.6
2
0.4
8
10.7
1
0.76
3.3
7
8.6
5
0.6
7
0.3
9
DO
L
2.2
6
HM
L
7.6
2
7.6
1
0.3
4
0.3
0
6.7
5
0.3
7
1.1
3
DO
L
HM
L
0.78
2.1
2
6.6
6
0.1
9
5.3
7
9.7
4
1.3
7
0.53
0.3
2
0.5
5
0.57
5.65
10.7
4
161
1
0.8
2
9.9
8
0.0
5
0.0
8
1.35
10.1
3
0.21
3.0
5
9.9
7
0.0
3
2.8
1
9.7
8
0.4
3
3.42
10.2
4
6.19
10.4
8
0.92
0.71
0.3 0.2
0.13 1
9
0.33 0.59
Log return (with b-a)
2
3
4
5
6
0.5
6
9.9
8
1.11
10.1
1
0.0
5
0.0
6
0.21
2.6
4
9.9
7
0.0
3
2.3
6
9.8
5
0.4
3
0.11
0.2
6
0.2
4
DO
L
HM
L
2.6
7
9.2
7
0.3
1
7.0
1
7.7
9
1.1
3
0.2
9
0.9
0
DO
L
HM
L
6.1
1
7.6
3
1.1
3
0.8
0
3.15
10.0
6
5.55
10.4
7
0.92
0.71
2.3
7
9.2
4
0.3
1
0.31
0.53
0.2
6
1.3 2.1
2.5
8
0
9
Std.
8.4 8.2
8.7
Dev.
8
7
5
Skewne
ss
0.0
0.3 0.2
0.5
1
5
1
9
SR
0.2
0.1 0.2
0.3
4
6
5
0
Discrete returns (with b-a)
Portfoli 1 2
3
4
o
Mean
1.7
1.7 1.6
2.0
2
6
8
1
Std.
8.4
8.4 8.2
8.7
Dev.
5
8
6
5
Skewne
ss
0.0
0.3 0.2
0.6
1
6
1
1
SR
0.2
0.2 0.2
0.2
0
1
0
3
3.5
2
8.7
4
0.7
9
2.7
9
8.7
7
0.8
4
0.3
2
HML
1.21
1.6
4
7.6
4
0.3
4
6.9
7
6.7
9
0.3
7
0.44
0.2
1
1.0
3
4.98
11.2
1
0.4
0
5
DO
L
DO
L
HML
1.23
1.2
5
7.6
4
0.4
7
6.2
3
6.8
4
0.4
4
0.40
0.1
6
0.9
1
4.51
11.2
4
1
1.4
9
9.9
6
0.0
5
0.1
5
1
1.2
4
9.9
6
0.0
5
0.1
2
0.76
10.1
5
0.32
2.5
3
9.9
7
0.1
4
2.1
2
9.9
2
0.6
0
2.74
10.4
1
5.38
10.6
7
1.12
0.92
0.2 0.2
0.07 5
1
0.26 0.50
Discrete returns (with b-a)
2
3
4
5
6
0.33
2.1
2
9.9
6
0.1
4
1.9
0
9.8
5
0.5
8
0.05
0.2
1
0.1
9
0.52
10.1
3
DO
L
HM
L
2.0
1
9.3
2
0.4
3
6.8
7
7.8
8
1.2
4
0.2
2
0.8
7
DO
L
HM
L
5.9
8
7.6
8
0.8
7
0.7
8
2.48
10.2
0
4.74
10.6
7
0.83
0.91
1.7
5
9.2
7
0.3
9
0.24
0.44
0.1
9
Following Lustig et al (2011) we use the DOL factor as the currency market excess return. DOL is the average
return of borrowing the US dollar and lending other currencies available in each sample. From Table 2, the
DOL factors are about 2.3% and 2.7% for two samples respectively, but it reduced to 2. 1% and 2.4%
respectively when taking transaction cost into account, which suggests that US investors demand a low but
positive risk premium for holding foreign currency. We use two ways to proxy the FX market volatility, the
realized volatility generated by aggregating average daily changes in exchange rates into monthly observation
(following Menkhoff et al (2011)) and the conditional volatility from the Component-GARCH model.(following
Adrian and Rosenberg (2008)).
The realized volatility: following Menkhoff et al (2011)), we define the volatility innovation factor as VOL,
and it is defined in Eq. (3)
Eq.(3)
1
where
(= ) is the absolute daily log return for each currency k on each day in the sample. Then
average over all currencies available on any given day and average daily values up to the monthly frequency.
This proxy is similar to measures of realized volatility (Andersen, Bollerslev, Diebold, and Labys (2001)).
They use absolute returns and not squared returns in order to minimize the impact of outliers. In the
empirical analysis, they focus on volatility innovations , as a non-traded risk factor. Since the first
differences are significantly autocorrelated, therefore they estimate a simple AR (1) for the volatility level and
take the residuals as main proxy for innovations. This factor is shown on Figure 2.
The conditional volatility: We apply the component-GARCH model to obtain the conditional volatilities at
different frequencies. The input of the estimation is the time series of montly FX market return, which we use
DOL factor as a proxy. The model and estimation of the component-GARCH model are shown in the next
section.
162
Global FX volatility
Global FX Volatility
.016
.016
.012
.012
.008
.008
.004
.004
.000
.000
-.004
-.004
84
86
88
90
92
94
96
98
00
02
04
06
08
10
Year
Volatility
Volatility innovations
2. Methodology
Modeling Conditional Volatilities
Specification of Volatility Dynamics: We follow Adrian and Rosenberg (2008) specification of the
Component-GARCH model. They use the conditionally log-normal models of volatility as Nelson (1991) shows
that conditionally log-normal models of volatility perform better than square-root or affine volatility
specifications. In modeling FX market risk, we incorporate these features and specify the dynamics of the
M
market excess return RX t and its conditional volatility as:
Market excess return:
(Eq.4)
RX M t 1 s l
1
2 t 1
3 t 1
t 1 t 1
Market volatility:
ln t 1 st 1 l t 1
(Eq.5)
Short-run component:
st 1 4 st 5 t 6 ( t 2 / )
(Eq. 6)
Long-run component:
lt 1 7 8lt 9t 10( t
2 / ) (Eq. 7)
where RX t 1 is the market excess return and it is proxy by the DOL factor mentioned in previous section.
The log-volatility in (Eq.4) is the sum of two components. Each component is a first order autoregressive
process AR(1) with its own rate of mean reversion. Without loss of generality, let l be the slowly meanM
The term
2 / in (Eq.6) and (Eq.7) are the shocks to the volatility component. For these error
terms, equal-sized positive or negative innovations result in the same volatility changes, although the
magnitude can be different for the short and long run components. We also allow for an asymmetric effect of
returns on short- and long-run volatilities by including the market innovation in (Eq.6) and (Eq.7). is a
normal i.i.d. error term with zero expectation and unit variance.
Equilibrium Asset Pricing Restrictions: Expected returns are endogenous. The key insight of intertemporal
equilibrium models such as Mertons (1973) ICAPM is that state variables of the return generating process
are state variables of the pricing kernel. In our setting with a two-component volatility process, the
163
+1
= +1
, +1
+ +1
, +1 + (+1
, +1 )
(Eq.8)
where is the coefficient of relative risk aversion, and and are (negatively) proportional to changes in
the marginal utility of wealth due to changes in the state variables and .
Eq. (8) shows that expected returns depend on three risk premiums. The first risk premium arises from the
covariance of the asset return with the market return, multiplied by relative risk aversion . This is the riskreturn trade-off in a static CAPM model. The second and third risk premiums depend on the covariance of the
asset return with the innovations in the short- and long-run factors. These are scaled by the impact of
changes in the volatility factors on marginal utility of wealth. 9 We allow for time-varying relative risk
aversion in equation (2) to accommodate non-separable preferences. Eq. (8) can be derived in Duffie and
Epsteins (1992) economy with stochastic differential utility. i In this intertemporal pricing framework, the
evolution of market volatility is exogenous. Thus, market volatility in our setup can be linked to the variability
of fundamental economic variables (for example, Cox, Ingersoll, and Ross (1985) and Tauchen (2005) address
the equilibrium pricing of this type of risk). Market volatility can also reflect uncertainty generated
endogeonously by financial constraints as suggested by Cuoco (1997) as well as Detemple and Serrat (2003).
Estimation of the Volatility Components: We estimate this volatility model using monthly market excess
returns, the DOL factor. It is the average currency excess return of a large set of currencies (currencies for 48
countries, depends on availability) against the USD. We estimate the volatility model from 12/1983 to
09/2011. Summary statistics for the monthly market excess return are given in the first row of table 3, and
Maximum Likelihood estimation results for the volatility model are shown in the remaining rows of the table.
In the expected return equation, we find that short-run volatility has a significant negative coefficient (2 ),
while the long-run volatility has a significant positive coefficient (3 ). The expected return thus depends
positively on long-run volatility but negatively on short-run volatility.
We can identify the short-and long-run components by their relative degrees of autocorrelation: the shortrun volatility component has an autoregressive coefficient (4 ) of 0.311, and the long-run component has an
autoregressive coefficient of (8 ) 0.976. While the long-run component is highly persistent, it is not
permanent; we reject the hypothesis that 8 = 1 at 1% level. Because the short-and long-run components
determine log-volatility additively, we are not able to identify the means of the two components separately,
and we estimate only the mean of the long-run component (7 ). We find negative returns increase short run
volatility more than positive returns. The asymmetric effect for the short-run component is large and
significant, while the asymmetric effect for the long-run component is not significant. Thus, we would expect
short-run volatility to be closely linked to market skewness, since a negative return shock disproportionately
increases short-run volatility, which further raises the likelihood of another large move. In Figure 3, we plot
two measures of the annualized volatility of the market return at a monthly frequency for 12/1983 to
09/2011. The first measure is the VOL factor used in Menkhoff et al (2011) as a proxy for FX market volatility;
it is similar to realized volatility. The second measure is the conditional volatility from our volatility
components model as shown in Table 3. It is apparent that they are quite similar. In Figure 4 we plot the
estimated short-run volatility component . Figure 5 plots the time series of the short-run volatility
component , which is clearly much less persistent than the long-run component.
Table 3: Time-Series Estimation of the Volatility Components, Monthly 12/1983 to 09/2011
This table reports the summary statistics of the monthly market excess return and the maximum likelihood
estimates of the volatility components model. The market excess return is measured as the average monthly
9
Merton (1973) and Cox, Ingersoll, and Ross (1985) derive general results that allow us to write expression in
equation (3.5)
164
Kurtosis
0.9300
3
0.0139***
0.0000
0.0527***
0.0034
10
-0.0886***
0.9761***
0.0015
0.0545***
0.0001
0.0000
0.0001
0.0012
0.0030
Ljung-Box Q-statistic of t
p-value
10 lags
11.30
20 lags
18.80
0.33
0.54
Figure 3: Monthly market volatility (annualized): This figure plots two measures of the annualized
volatility of the market return at a monthly frequency for 12/1983 to 09/2011. The first measure is the VOL
factor used in Menkhoff et al (2011), it is a measure of realized volatility. The second measure is the
conditional volatility from the volatility components model shown in table 3. Shaded areas correspond to
NBER recessions10.
VOLO
.014
.012
.010
.008
.006
.004
.002
.000
1985
1990
1995
2000
2005
2010
2000
2005
2010
V
.0016
.0014
.0012
.0010
.0008
.0006
.0004
.0002
1985
1990
1995
There are three NBER recessions covered in our sample period: July 1990 (III) to March 1991 (I), March 2001 (I) to
November 2001 (IV) and December 2007 (IV) to June 2009 (II)
10
165
Long-run component
-3.6
-3.7
-3.8
-3.9
-4.0
-4.1
-4.2
84
86
88
90
92
94
96
98
00
02
04
06
08
10
Year
Figure 5 the short-run volatility component This figure plots the estimated short-run volatility
component () at a monthly frequency from November 1983 to September 2011. The estimate of is from the
stochastic volatility model that is reported in table 3.1.
.5
Short-run component
.4
.3
.2
.1
.0
-.1
84
86
88
90
92
94
96
98
00
02
04
06
08
10
Year
Asset Pricing Test: This section summarizes our approach to cross-sectional asset pricing. The benchmark
results rely on a standard stochastic discount factor approach (Cochrane, 2005), which also used in Menkhoff
et al. (2011) and in Lustig et al. (2011). Following Lustig et al. (2011) and Menkhoff et al. (2011), this paper
uses discrete returns (not log returns) in all pricing exercise to satisfy the Euler equation. Discrete returns for
+1
where F and S are the level of the forward and spot exchange rate
respectively. The descriptive statistics of the discrete returns for portfolios are provided in Table 2. The carry
trade is a zero-cost trading strategy and no-arbitrage relation applies so that risk-adjusted currency excess
returns have a zero price and satisfy the basic Euler equation:
Eq.(8)
+1 +1
=0
Here, +1 denotes the SDF that prices returns denominated in dollars. The unconditional version of Eq. (8)
is:
= 0
Eq.(9)
This equation can be written as:
+ , = 0
Eq.(10)
In practice, the average unconditional returns to the strategies that we consider are positive. The most
straightforward explanation of this finding is that , < 0.
Our analysis uses Eq.(9) as our point of departure. We consider linear SDFs that take the form
= 1 ( )
Eq.(11)
166
167
168
lambda 0.110
-0.061
9.119
(ols)
0.122
0.018
<0.058>
<Sh>
0.122
0.021
[HAC]
0.131
0.027
Panel B: Factor Betas
All-country
PF
Alpha
DOL
VOL
1
-0.002
0.974
4.18
[0.001]
[0.047]
[0.751]
2
-0.002
0.932
0.672
[0.001]
[0.042]
[0.799]
J-stats
1.084
0.781
6.9e-004
6.967
[0.138]
2
0.764
0.714
OECD-country
GMM
DOL%
B
-0.012
s.e.
(0.033)
lambda
0.150
s.e.
(0.146)
FMB
DOL%
lambda
0.150
(ols)
0.143
<Sh>
0.143
[HAC]
0.148
VOL%
-4.879
(2.524)
-0.050
(0.021)
VOL%
-0.050
0.017
0.018
0.023
2
0.784
J-stats
5.394
0.249
MAE
6.001
<0.199>
6.2e-004
4.187
[0.381]
OECD-country
PF
Alpha
1
-0.003
[0.001]
2
-0.001
[0.001]
DOL
0.986
[0.044]
1.048
[0.036]
V0L
4.491
[0.996]
1.374
[0.667]
2
0.785
169
0.869
0.000
[0.001]
0.001
[0.001]
0.001
[0.001]
0.003
[0.001]
0.976
[0.036]
1.028
[0.039]
0.99
[0.042]
1.101
[0.076]
1.302
[0.611]
-0.703
[0.645]
-1.021
[0.856]
-4.429
[1.372]
0.815
0.835
0.777
0.634
0.000
[0.001]
0.000
[0.001]
0.001
[0.001]
0.003
[0.001]
1.035
[0.028]
1.004
[0.029]
1.007
[0.031]
0.921
[0.059]
1.687
[0.676]
-1.37
[0.604]
-2.63
[0.773]
-3.553
[1.233]
0.874
0.874
0.839
0.659
Table 5 Summary Statistics of Pricing Factors: The table reports mean, standard deviations (the mean and
standard deviation are in percentage), skewness and kurtosis for 6 different factors. The DOL and VOL factor
are generated simply from the data and defined in section 3, while the other four factors are generated from
the C-GARCH model. The V factor is the conditional market volatility measure, and the VRES is the market
volatility innovation which is generated by the AR(2) residual of factor V. The LRES and SRES are long-run and
short run volatility innovations. All are monthly observations and the time period is spanned from
01/11/1983 to 30/09/2011.
Pricing Factor
DOL
VOL
V
Market variance (vres)
Short-run volatility (sres)
Long-run volatility (lres)
Mean (%)
0.1480
0.0009
0.0500
-0.0000
-0.0003
-0.0139
SD (%)
2.262
0.1018
0.0190
0.0132
7.5356
3.4163
Skewness
-0.5175
1.4981
1.4310
2.5315
1.9321
1.1414
Kurtosis
1.1363
5.1793
2.7009
8.1394
3.8489
0.9984
Table 6. Factor Loadings of the 5 forward discount sorted portfolios for 48-all-country sample: This
table has the same design as Table 4 panel (a) and it reports factor loadings from regressions of each portfolio
return on the DOL factor, long-run volatility innovations (LRES), and short-run volatility innovations (SRES)
(The time-series Fama and MacBeth (1973) regressions) Standard errors are adjusted for heteroskedasticity.
Panel (a) are the results for first stage of regression (5) in Table 8, Panel (b) are the results for first stage of
regression (6) in Table 8 and Panel (c) are the results for first stage of regression (7) in Table 8. The sample
period is spanned from 01/11/1983 to 30/09/2011.
Factor Betas for 48-all-country sample
Panel (a) 1st stage of Regression (5)
P
Alph
DOL
LRES
SRES
F
a
1
0.945
0.051
0.009
0.00
2
[0.051 [0.023 [0.011
0.00
]
]
]
1
2
0.941
0.005
0.034
0.00
2
0.00
[0.040 [0.023 [0.011
1
]
]
]
3
0.00
0.969
-0.035 0.017
0
[0.036 [0.028 [0.012
0.00
]
]
]
1
4
0.00
1.028
-0.007 -0.011
1
[0.037 [0.020 [0.010
0.00
]
]
]
1
5
0.00
0.997
-0.009 -0.002
1
0.00
[0.044 [0.023 [0.011
2
0.74
2
0.72
5
0.81
6
0.83
6
0.77
6
170
0.00
1
0.00
2
0.00
1
0.00
0
0.00
1
0.00
1
0.00
1
0.00
1
0.00
[0.051
]
0.941
[0.040
]
0.968
[0.037
]
1.028
[0.037
]
0.997
[0.044
[0.011
]
0.035
[0.009
]
0.009
[0.009
]
-0.013
[0.009
]
-0.004
0.72
5
0.81
5
0.83
6
0.77
6
0.00
3
1.120
-0.006
-0.047
[0.082
]
[0.037
]
[0.020
]
0.00
1
0.62
8
0.00
3
1.135
-0.059
[0.087
]
[0.039
]
0.00
1
0.61
9
0.00
3
1.120
0.00
1
[0.082
]
[0.009
]
-0.049
0.62
8
[0.019
]
Table 7 Factor Loadings of the 5 forward discount sorted portfolios for 29-OECD-country Sample: This
table has the same design as Table 4 panel (a) and it reports factor loadings from regressions of each portfolio
return on the DOL factor, long-run volatility innovations (LRES), and short-run volatility innovations (SRES)
(The time-series Fama and MacBeth (1973) regressions) Standard errors are adjusted for heteroskedasticity.
Panel (a) are the results for first stage of regression (5) in Table 9, Panel (b) are the results for first stage of
regression (6) in Table 9 and Panel (c) are the results for first stage of regression (7) in Table 9. The sample
period is spanned from 01/11/1983 to 30/09/2011.
Factor Betas for 29-OECD-country sample
Panel (a) 1st stage of Regression (5)
P
Alph
DOL
LRES
SRES
R2
F
a
1
0.00
0.76
2
0.963
0.026
0.015
5
0.00
1
0.00
1
[0.053
]
[0.026
]
[0.013
]
1.045
0.021
0.016
0.00
1
0.00
0
[0.034
]
[0.018
]
[0.010
]
1.025
0.025
-0.001
0.00
0
0.00
0
[0.032
]
[0.022
]
[0.011
]
1.013
0.001
-0.000
0.00
1
0.00
1
0.00
1
0.00
3
[0.028
]
[0.022
]
[0.009
]
1.021
[0.032
]
-0.032
[0.025
]
-0.005
[0.011
]
0.934
-0.041
-0.025
0.00
1
[0.065
]
[0.033
]
[0.019
]
0.87
2
0.00
1
0.00
1
1.041
0.039
0.87
1
0.00
1
0.00
1
[0.034
]
[0.019
]
1.025
0.024
0.00
1
0.00
0
[0.031
]
[0.021
]
1.013
0.001
0.00
0
0.00
1
0.00
1
0.00
3
[0.028
]
[0.018
]
1.022
[0.032
]
-0.037
[0.020
]
0.94
-0.069
0.00
1
[0.066
]
[0.039
]
0.87
2
0.83
2
0.65
4
[0.025
]
0.86
9
0.00
1
0.00
1
0.87
1
0.00
1
0.00
0
0.87
2
0.83
2
0.65
2
0.00
1
0.00
0
0.00
1
0.00
1
0.00
1
0.00
3
0.00
1
[0.053
]
1.045
[0.035
]
1.025
[0.031
]
1.013
[0.028
]
1.02
[0.033
]
0.933
[0.065
]
[0.012
]
0.021
[0.009
]
0.005
[0.010
]
0.000
[0.007
]
-0.012
[0.008
]
-0.034
0.87
0
0.87
0
0.87
2
0.83
1
0.65
3
[0.020
]
Table 8 Factor prices of different factors for 48-all-country sample: This Table reports summary
statistics of the 2nd stage (cross-sectional) Fama and MacBeth (1973) regressions for the forward discount
sorted portfolios of the 48-all-country sample. In the first stage, portfolio returns are regressed on the pricing
factors to obtain factor loadings (reported in table 6). In the second stage, for each month, portfolio returns
are regressed on the loadings, giving an estimate of the price of risk for each factor. The t-statistics are
adjusted for autocorrelation and heteroskedasticity. (Significance at the 1% level is denoted by ***, at the 5%
level by **, and at the 10% level by *)
171
Coef.
SE(sh)
Coef.
SE(sh)
Short-run volatility
(SRES)
Coef.
SE(sh)
Market variance
(VERS)(%)
Coef.
SE(sh)
HML
Coef.
SE(sh)
Coef.
SE(sh)
VOL (%)
MAE
(1)
0.0012
0.0012
J-statistics
(3)
0.0015
0.0012
(4)
0.0012
0.0012
(5)
0.0015
0.0012
0.0389**
0.0171
0.0552**
0.028
(6)
0.0011
0.0011
0.0441***
0.0180
0.9580
2.9725e004
<0.873>
0.9500
3.6079e004
<0.935>
(7)
0.0011
0.0010
0.0630***
0.024
0.0140***
0.0064
0.0052***
0.0015
0.061***
0.024
0.0730
0.0016
(2)
0.0011
0.0012
0.002
0.7353
9.3119e004
<0.079>
0.732
6.9000e004
0.781
0.9274
4.4621e004
<0.841>
0.8462
6.4937e004
<0.354>
Table 9 Factor prices of different factors for 29-OECD-country sample: This Table reports summary
statistics of the 2nd stage (cross-sectional) Fama and MacBeth (1973) regressions for the forward discount
sorted portfolios of the 29-oecd-country sample. In the first stage, portfolio returns are regressed on the
pricing factors to obtain factor loadings (reported in table 7). In the second stage, for each month, portfolio
returns are regressed on the loadings, giving an estimate of the price of risk for each factor. The t-statistics
are adjusted for autocorrelation and heteroskedasticity. (Significance at the 1% level is denoted by ***, at the
5% level by **, and at the 10% level by *)
Coef.
SE(sh)
Coef.
SE(sh)
Short-run volatility
(SRES)
Coef.
SE(sh)
Market variance
(VERS)(%)
Coef.
SE(sh)
HML
Coef.
SE(sh)
Coef.
SE(sh)
VOL (%)
MAE
J-statistics
(1)
0.0015
0.0014
(2)
0.0015
0.0014
(3)
0.0015
0.0014
(4)
0.0018
0.0014
(5)
0.0015
0.0011
-0.0006
0.0036
(6)
0.0015
0.0012
0.0336**
0.0160
-0.1087
0.0360
(7)
0.0015
0.0012
0.0758**
0.0360
0.0122**
0.0048
0.0048***
0.0016
0.0050***
0.0211
0.0114
0.0012
<0.032>
0.9342
3.0334e004
<0.633>
172
0.784
6.2142e004
<0.249>
0.8633
4.1411e004
<0.819>
0.9222
3.0149e004
<0.853>
0.8038
5.5523e004
<0.660>
0.8928
3.9953e004
<0.901>
All countries
OECD Countries
8
3
4
2
2
0
1
-2
-4
6
5
2 3
0
1
-2
-4
-4
-2
-4
-2
5. Conclusion
This paper is inspired by the application of ICAPM theory in FX market (Menkhoff et al. (2011)) and the
success of component GARCH model in cross-sectional pricing of stock market ( Adrian and Rosenberg
(2008)). We use equally weighted 48-currency portfolio as market portfolio, and decompose the return of
market portfolio in to long- and short-run volatility components according to the same volatility dynamic
specification of Adrian and Rosenberg (2011). We find that prices of risk are negative and significant (for allcountry sample) for both volatility components implies that investors pay for insurance against increases in
volatility, even if those increases have little persistence. As far as we know, although there is literature using
component GARCH model to decompose volatility returns for a single currency, we are the first to decompose
FX market volatility. Different from stock market, it is difficult to provide a unique measure for the currency
market return. The return from currency market has to be a return from certain trading strategy; however,
for different trading strategy the average returns are different. Without loss of general measure, we use DOL
factor of 48-all-country sample as a measure of market volatility, this measure is not perfect, because, it does
not/only take (1) 48 currencies and not all the currencies, (2) currencies are equally weighted (3) no specific
trading strategies have been taking into consideration. Therefore, this can explain why the decomposed
factors from the DOL measure works quite well for the 48-all-country sample but not very well for the 29OECD sample as the measure of FX market return is still not general enough to kill the currency specific
effects.
For further research, it would be interesting to study the two volatility components as what Adrian and
Rosenberg (2008) have done for the stock market. They find that their long-run volatility component is highly
correlated with business cycle risk while the short-run volatility component is correlated with market
sentiments. Currency market is a different story as the difference between those two markets. This coincides
with Burnside (2011) in which he tests all traditional factors used in stock market pricing to explain the
excess returns from cross-sectional currency portfolios and finds that none of them work. But since the two
173
174
Equation (3.2) can be derived in Duffie and Epsteins (1992) stochastic differential utility setting by
replacing their equation (17) into (36), imposing market clearing, and then solving for the expected returns of
individual assets. Adrian and Rosenberg (2008) page 3001 footnote 4 (prove this in appendix)
i
175