You are on page 1of 3

Denise Keaton

Bsg 502
CASE STUDY FOR CHAPTER 3
Demand Estimation for The San Francisco Bread Co.
Given the highly competitive nature of the restaurant industry, individual companies cautiously
guard operating information for individual outlets. As a result, there is not any publicly available
data that can be used to estimate important operating relationships. To see the process that
might be undertaken to develop a better understanding of store location decisions, consider the
hypothetical example of The San Francisco Bread Co., a San Francisco-based chain of bakerycafes. San Francisco has initiated an empirical estimation of customer traffic at 30 regional
locations to help the firm formulate pricing and promotional plans for the coming year. Annual
operating data for the 30 outlets appear in Table 3.5.The following regression equation was fit to
these data:Qi= b0+ b1 P i+ b2 P xi+ b3Ad i+ b4I i+ u it Q is the number of meals served, P is the
average price per meal (customer ticket amount, in dollars), P xis the average price charged by
competitors (in dollars), Ad is the local advertising budget for each outlet (in dollars), I is the
average income per household in each outlets immediate service area, and ui
is a residual (or disturbance) term. The subscript i indicates the regional market from which the
observation was taken. Least squares estimation of the regression equation on the basis of the
30 data observations resulted in the estimated regression coefficients and other statistics shown
in Table 3.6.
A.Describe the economic meaning and statistical significance of each individual independent
variable included in the San Francisco demand equation.
B.Interpret the coefficient of determination (R2) for the San Francisco demand equation.
C.What are expected unit sales and sales revenue in a typical market?
D.To illustrate use of the standard error of the estimate statistic, derive the 95 percent
confidence interval for expected unit sales and total sales revenue in a typical market.
CASE STUDY SOLUTION
A.Individual coefficients for the San Francisco Bread Co. regression equation can be interpreted
as follows. The intercept term, 128,832.240, has no economic meaning in this instance; it lies far
outside the range of observed data and obviously cannot be interpreted as the expected unit
sales of a given San Francisco Bread outlet when all the independent variables take on zero
values. The coefficient for each independent variable indicates the marginal relation between
that variable and unit sales, holding constant the effects of all the other variables in the demand
function. For example, the -19,875.954 coefficient for P, the average price charged per meal

(customer ticket amount), indicates that when the effects of all other demand variables are held
constant, each $1 increase in price causes annual sales to decline by 19,875.954 units.The
15,467.936 coefficient for Px, the competitor price variable, indicates that demand for San
Francisco meals rises by 15,467.936 units per year with every $1 increase in competitor prices.
Ad, the advertising and promotion variable, indicates that for each$1 increase in advertising
during the year, an average of 0.261 additional meals are sold. The 8.780 coefficient for the I
variable indicates that, on average, a $1 increase in the average disposable income per
household for a given market leads to a8.780-unit increase in annual meal demand.Individual
coefficients provide useful estimates of the expected marginal influence on demand following a
one-unit change in each respective variable.However, they are only estimates. For example, it
would be very unusual for a $1increase in price to cause exactly a -19,875.954-unit change in
the quantity demanded.The actual effect could be more or less. For decision-making purposes,
it would be helpful to know if the marginal influences suggested by the regression model are
stable or instead tend to vary widely over the sample analyzed.In general, if it is known with
certainty that Y = a + bX , then a one-unit change in X will always lead to ab-unit change in Y. If
b> 0, X andY will be directly related;if b< 0, X andY will be inversely related. If no relation at all
holds between X and Y,then b = 0. Although the true parameter b is unobservable, its value is
estimated by the regression coefficient b. If b= 10, a 1-unit change in X will increaseY
by 10 units.This effect may appear to be large, but it will be statistically significant only if it is
stable over the entire sample. To be statistically reliable,b must be large relative to its degree of
variation over the sample.In a regression equation, there is a 68 percent probability that b lies in
the interval b one standard error (or standard deviation) of b. There is a 95 percent
probability that b lies in the interval b two standard errors of b. There is a 99 percent
probability that b is in the interval b three standard errors of b. When a coefficient is at least
twice as large as its standard error, one can reject at the 95 percent confidence level the
hypothesis that the true parameter b equals zero. This leaves only a 5 percent chance of
concluding incorrectly that b0 when in fact b = 0.When a coefficient is at least three times as
large as its standard error (standard deviation), the confidence level rises to 99 percent and the
chance of error falls to 1 percent.A significant relation between X and Y is typically indicated
whenever a coefficient is at least twice as large as its standard error; significance is even more
likely when a coefficient is at least three times as large as its standard error. The independent
effect of each independent variable on sales is measured using a two-tail t statistic where: t
statistic =b /Standard error of b This t statistic is a measure of the number of standard errors
between b and a hypothesized value of zero. If the sample used to estimate the regression
parameters is large (for example,n> 30), the t statistic follows a normal distribution and
properties of a normal distribution, can be used to make confidence statements concerning the
statistical significance of b. Hence t = 1 implies 68 percent confidence, t = 2 implies95 percent
confidence,t = 3 implies 99 percent confidence, and so on. For small sample sizes (for example,
d.f. = n - k < 30), the t -distribution deviates from a normal distribution, and a t table should be
used for testing the significance of estimated regression parameters. In San Francisco Breads
demand equation, each coefficient estimate is more than twice as large as its standard error.

Therefore, it is possible to reject the hypothesis that each of the independent variables is
unrelated to demand with95 percent confidence. The coefficients suggest an especially strong
relation between demand and the P, P x, and I variables. Each of these coefficients is over
three times as large as its underlying standard error and therefore is statistically significant at
the 99 percent confidence level.
B.
The coefficient of determination ( R2) indicates that the regression model has explained83.3
percent of the total variation in demand. This is a very satisfactory level of explanation for the
model as a whole.
C.
To project annual unit sales for a typical market, the company must simply enter expected
values for each independent variable in the estimated demand equation. SanFrancisco Bread
expects a typical average price of $6.93, competitor price of $6.16,advertising of $244,649, and
household income of $51,044. Inserting the appropriate unit values into the demand equation
results in estimated demand of:
Q=128,832.240 - 19,875.954(6.93) + 15,467.936(6.16) + 0.261(244,649)+
8.780(51,044)=598,394.074 units (meals).
Then, expected total revenue can be estimated by multiplying estimated unit demand by the
typical sales price of $6.93:
TR=P Q=598,394.074 $6.93
=$4,146,870.93.
D.
Although 598,394.074 units is the best estimate of demand for a typical market, it ishighly
unlikely that precisely this number of meals will actually be served. Either more or less may be
sold, depending on the effects of other factors not explicitly accounted for in this demand
estimation model. The standard error of the estimate is avery useful statistic because it allows
us to construct a range or confidence interval within which actual unit sales are likely to fall. For
example, sales can be projected to fall within a range of 2 standard errors of the 598,394.074
units expected sales level with a confidence level of 95 percent. The standard error of the
estimate for the SanFrancisco demand-estimation model is roughly 14,875.95. Therefore, an
interval of 2(14,875.95) or 29,751.9 units around the expected unit sales of
598,394.074represents the 95 percent confidence interval. This means that one can predict with
a95 percent probability of being correct that unit sales at a typical outlet will lie in the range from
568,642.174 to 628,145.974 units (meals). Assuming that the hypothesized values for the
various independent variables are correct, there is only a 5 percent chance that actual unit sales
will fall outside this range. Similarly, a 95 percent probability of being correct that total revenue at
a typical outlet will lie in the range from $3,940,690.27 (= 568,642.174 $6.93) t $4,353,051.60
(= 628,145.974 $6.93).

You might also like