You are on page 1of 11

options ls=76 ps=100;

data bin;

input presion resp ind;

n=1;

cards;

117 1 5

117 0 153

121.5 1 17

121.5 0 235

131.5 1 12

131.5 0 272

141.5 1 16

141.5 0 255

151.5 1 12

151.5 0 127

161.5 1 8

161.5 0 77

176.5 1 16

176.5 0 83

186 1 8

186 0 35

proc genmod data=bin;

model resp/n=presion/dist=binomial link=logit wald type1;

weight ind;

run;
The SAS System 1
09:47 Thursday, August 8, 2002

The GENMOD Procedure

Model Information

Data Set WORK.BIN


Distribution Binomial
Link Function Logit
Response Variable (Events) resp
Response Variable (Trials) n
Scale Weight Variable ind
Observations Used 16
Number Of Events 8
Number Of Trials 16

Criteria For Assessing Goodness Of Fit

Criterion DF Value Value/DF

Deviance 14 658.4571 47.0327


Scaled Deviance 14 658.4571 47.0327
Pearson Chi-Square 14 1344.2793 96.0199
Scaled Pearson X2 14 1344.2793 96.0199
Log Likelihood -329.2286

Algorithm converged.

Analysis Of Parameter Estimates

Standard Wald 95% Chi-


Parameter DF Estimate Error Confidence Limits Square Pr > ChiSq

Intercept 1 -6.0078 0.7591 -7.4955 -4.5200 62.64 <.0001


presion 1 0.0240 0.0051 0.0140 0.0340 22.06 <.0001
Scale 0 1.0000 0.0000 1.0000 1.0000

NOTE: The scale parameter was held fixed.

LR Statistics For Type 1 Analysis

Chi-
Source Deviance DF Square Pr > ChiSq

Intercept 679.4729
presion 658.4571 1 21.02 <.0001

options ls=76 ps=100;


data bin;
input larvas fus;
n=1;
cards;
9 0
13 0
6 0
8 0
10 0
4 0
14 0
8 0
11 0
7 0
9 0
7 0
5 0
14 0
13 1
16 1
10 1
12 1
11 1
14 1
15 1
18 1
7 1
16 1
9 1
9 1
11 1
28 1
29 1
30 1
31 1
32 1
33 1
34 1
35 1
36 1
37 1
38 1
39 1
40 1
41 1
42 1
43 1
44 1
45 1
46 1
47 1
48 1
49 1
50 1
51 1
52 1
53 1
54 1
;
proc genmod data=bin;
model fus/n=larvas/dist=binomial link=logit wald type1;
run;
The SAS System 4
09:47 Thursday, August 8, 2002

The GENMOD Procedure

Model Information

Data Set WORK.BIN

Distribution Binomial

Link Function Logit

Response Variable (Events) fus

Response Variable (Trials) n

Observations Used 54

Number Of Events 40

Number Of Trials 54

Criteria For Assessing Goodness Of Fit

Criterion DF Value Value/DF

Deviance 52 30.3761 0.5842

Scaled Deviance 52 30.3761 0.5842

Pearson Chi-Square 52 25.4075 0.4886

Scaled Pearson X2 52 25.4075 0.4886

Log Likelihood -15.1881


Algorithm converged.

Analysis Of Parameter Estimates

Standard Wald 95% Chi-

Parameter DF Estimate Error Confidence Limits Square Pr > ChiSq

Intercept 1 -3.6864 1.5714 -6.7663 -0.6064 5.50 0.0190

larvas 1 0.3408 0.1422 0.0621 0.6196 5.74 0.0166

Scale 0 1.0000 0.0000 1.0000 1.0000

NOTE: The scale parameter was held fixed.

LR Statistics For Type 1 Analysis

Chi-

Source Deviance DF Square Pr > ChiSq

Intercept 61.8063

larvas 30.3761 1 31.43 <.0001

options ls=76 ps=120;


data rm;
input vap vpn alpl lva gva oid fus p100 rend;
cards;

13.7 1.415 125 6.6 4.2 1 2 23 1469.72


11.9 1.21 100.5 7.35 4.9 1 2 26.65 1384.58
19.1 1.255 124 5.65 4.7 0.5 1 23.25 1832.08
9.8 1.25 107.8 7.7 3.9 1.5 1.5 25.8 993.61
19.5 1.3 119.7 6.85 4.3 1 1 22.75 1697.08
16 1.35 109.6 6.3 4.2 3 1 19.25 1255.83
8.2 1.22 97.5 7.2 5.4 2 1 21.75 1371.81
14.2 1.115 102.5 7.15 5.1 0.5 1 29.25 1539.31
19.8 1.405 123.7 7.4 4.6 1.5 1 30.7 1763.19
11.8 1.19 94.9 6.2 6.1 0.5 1 20 1990.93
8.7 1.325 85.6 7.05 4.4 2 2.5 26.85 1000.56
6.2 1.025 83.9 7.1 4.9 1 2 23.4 736.25
12.1 1.17 72.6 5.95 4.7 3.5 1.5 12.65 806.39
9.1 1.31 97.9 6.5 4.2 0.5 2 27.15 1244.86
16 1.11 84.7 7.5 4.7 3.5 2.5 28.5 435.56
11.7 1.565 104.2 6.1 5 0 2 20.7 1667.08
22.1 1.47 106.7 6.25 5.9 1.5 1.5 15.75 2091.81
14.3 1.235 84.5 6.35 4.4 3.5 1.5 18.55 1427.64
8.2 1.33 85 6.55 4.8 3 2 14.4 452.5
8.1 1.08 93.2 7.2 6 2.5 3 20.55 871.94
13.1 1.285 103 6.85 5.4 2 1.5 20.3 1725.14
15.2 1.45 97.5 5.75 4.9 2 3 15.35 582.08
8.5 1.115 93.2 6.8 4.7 2 2.5 24.7 862.78
6.3 1.115 66 7.75 5.2 3.5 2.5 23.75 693.89
8.8 1.785 88 6.14 4.5 0 2 20.4 1163.06
8 1.355 91.9 7.01 4.9 1 2 19.45 1040.28
11.8 1.375 99.9 7.15 4.8 2 1.5 20.8 1112.08
8 1.175 78.4 6.85 3.5 3.5 3 24.05 678.33
12.7 1.11 89.7 6.95 4.5 1 1 23.8 1475.56
11.7 1.22 115.3 7.4 6.4 4 1.5 18.65 1380.14
8.1 1.235 107.1 7.05 4.3 3 2 26.9 581.53
9 1.33 111.5 6.75 4.5 3.5 1 22.9 1654.72
7.8 1.125 101.7 7.3 5.2 2.5 2 25.05 1314.17
20.7 1.305 140 6.7 4.1 1 1 25.15 2057.08
16.1 1.305 115.5 7.1 5 0 2.5 27.65 790.97
10.9 1.385 94.8 6.65 3.4 1 0.5 24.2 1454.44
16.6 1.33 106.5 6.5 3.7 3.5 1.5 24.15 1576.11
15.3 1.34 118 6.25 3.9 1 1.5 25.25 1404.31
16.1 1.575 131.8 5.8 3.8 1 1 26.6 2074.03
24.5 1.455 118.2 5.7 3.2 2 1.5 26.7 1387.36
15.7 1.48 120.5 6.15 5.4 1 1 20.65 1372.36
20.6 1.325 107.5 6.1 5.3 2 1 17.7 2155.83
;
proc reg;
model rend=vap vpn alpl lva gva oid fus p100/selection=maxr;
run;

The SAS System 5


09:47 Thursday, August 8, 2002

The REG Procedure


Model: MODEL1
Dependent Variable: rend

Maximum R-Square Improvement: Step 1

Variable fus Entered: R-Square = 0.5297 and C(p) = 17.0979


Analysis of Variance

Sum of Mean
Source DF Squares Square F Value Pr > F

Model 1 4840576 4840576 45.06 <.0001


Error 40 4297078 107427
Corrected Total 41 9137653

Parameter Standard
Variable Estimate Error Type II SS F Value Pr > F

Intercept 2180.01864 140.61877 25819450 240.34 <.0001


fus -528.45433 78.72552 4840576 45.06 <.0001

Bounds on condition number: 1, 1


----------------------------------------------------------------------------

The above model is the best 1-variable model found.

Maximum R-Square Improvement: Step 2

Variable alpl Entered: R-Square = 0.6228 and C(p) = 8.1973

Analysis of Variance

Sum of Mean
Source DF Squares Square F Value Pr > F

Model 2 5690715 2845358 32.19 <.0001


Error 39 3446938 88383
Corrected Total 41 9137653

Parameter Standard
Variable Estimate Error Type II SS F Value Pr > F

Intercept 931.34704 422.33312 429815 4.86 0.0334


alpl 10.21552 3.29382 850140 9.62 0.0036
fus -406.70304 81.48682 2201653 24.91 <.0001

Bounds on condition number: 1.3022, 5.2089


----------------------------------------------------------------------------

The above model is the best 2-variable model found.

Maximum R-Square Improvement: Step 3

Variable gva Entered: R-Square = 0.6534 and C(p) = 6.6112

Analysis of Variance

Sum of Mean
Source DF Squares Square F Value Pr > F

Model 3 5970391 1990130 23.88 <.0001


Error 38 3167262 83349
Corrected Total 41 9137653
Parameter Standard
Variable Estimate Error Type II SS F Value Pr > F

Intercept 319.33800 528.99051 30374 0.36 0.5496


alpl 10.91299 3.22122 956636 11.48 0.0017
gva 116.64707 63.67902 279676 3.36 0.0748
fus -410.61583 79.16100 2242586 26.91 <.0001

Bounds on condition number: 1.3207, 10.942


----------------------------------------------------------------------------

The above model is the best 3-variable model found.

Maximum R-Square Improvement: Step 4

Variable lva Entered: R-Square = 0.6842 and C(p) = 4.9954

Analysis of Variance

Sum of Mean
Source DF Squares Square F Value Pr > F

Model 4 6252391 1563098 20.04 <.0001


Error 37 2885262 77980
Corrected Total 41 9137653

Parameter Standard
Variable Estimate Error Type II SS F Value Pr > F

Intercept 1318.03961 733.22155 251982 3.23 0.0804


alpl 10.08916 3.14572 802146 10.29 0.0028
lva -156.13306 82.10369 281999 3.62 0.0650
gva 137.81669 62.59185 378052 4.85 0.0340
fus -390.60088 77.28898 1991659 25.54 <.0001
The SAS System 6
09:47 Thursday, August 8, 2002

The REG Procedure


Model: MODEL1
Dependent Variable: rend

Maximum R-Square Improvement: Step 4

Bounds on condition number: 1.3462, 19.411


----------------------------------------------------------------------------

The above model is the best 4-variable model found.

Maximum R-Square Improvement: Step 5

Variable p100 Entered: R-Square = 0.7020 and C(p) = 4.9124

Analysis of Variance

Sum of Mean
Source DF Squares Square F Value Pr > F

Model 5 6414840 1282968 16.96 <.0001


Error 36 2722813 75634
Corrected Total 41 9137653

Parameter Standard
Variable Estimate Error Type II SS F Value Pr > F

Intercept 1444.07991 727.20959 298249 3.94 0.0547


alpl 7.66905 3.51066 360929 4.77 0.0355
lva -256.88002 106.13124 443088 5.86 0.0207
gva 201.22220 75.31020 539957 7.14 0.0113
fus -404.97571 76.74667 2105980 27.84 <.0001
p100 23.03788 15.71961 162449 2.15 0.1515

Bounds on condition number: 2.2755, 44.322


----------------------------------------------------------------------------

The above model is the best 5-variable model found.

Maximum R-Square Improvement: Step 6

Variable vap Entered: R-Square = 0.7098 and C(p) = 6.0067

Analysis of Variance

Sum of Mean
Source DF Squares Square F Value Pr > F

Model 6 6485473 1080912 14.26 <.0001


Error 35 2652180 75777
Corrected Total 41 9137653

Parameter Standard
Variable Estimate Error Type II SS F Value Pr > F

Intercept 1227.60201 761.64813 196853 2.60 0.1160


vap 12.73335 13.18881 70633 0.93 0.3409
alpl 5.89442 3.96568 167410 2.21 0.1461
lva -220.21017 112.81716 288708 3.81 0.0590
gva 198.93719 75.41845 527244 6.96 0.0124
fus -393.54276 77.72649 1942588 25.64 <.0001
p100 22.08657 15.76527 148727 1.96 0.1700

Bounds on condition number: 2.2845, 70.184


----------------------------------------------------------------------------

The above model is the best 6-variable model found.

Maximum R-Square Improvement: Step 7

Variable vpn Entered: R-Square = 0.7162 and C(p) = 7.2484

Analysis of Variance

Sum of Mean
Source DF Squares Square F Value Pr > F

Model 7 6544612 934945 12.26 <.0001


Error 34 2593041 76266
Corrected Total 41 9137653

Parameter Standard
Variable Estimate Error Type II SS F Value Pr > F

Intercept 586.17847 1055.66466 23515 0.31 0.5823


vap 13.84554 13.29147 82757 1.09 0.3049
vpn 327.71675 372.15741 59139 0.78 0.3847
alpl 4.54886 4.26182 86885 1.14 0.2933
lva -185.19873 119.96124 181771 2.38 0.1319
gva 212.64116 77.24544 577936 7.58 0.0094
fus -400.60641 78.38856 1991870 26.12 <.0001
p100 24.37901 16.02891 176423 2.31 0.1375

Bounds on condition number: 2.5265, 99.189


----------------------------------------------------------------------------

The above model is the best 7-variable model found.

Maximum R-Square Improvement: Step 8


The SAS System 7
09:47 Thursday, August 8, 2002

The REG Procedure


Model: MODEL1
Dependent Variable: rend

Maximum R-Square Improvement: Step 8

Variable oid Entered: R-Square = 0.7183 and C(p) = 9.0000

Analysis of Variance

Sum of Mean
Source DF Squares Square F Value Pr > F

Model 8 6563988 820498 10.52 <.0001


Error 33 2573665 77990
Corrected Total 41 9137653

Parameter Standard
Variable Estimate Error Type II SS F Value Pr > F

Intercept 725.91955 1103.73019 33736 0.43 0.5153


vap 15.03682 13.65170 94619 1.21 0.2787
vpn 262.81665 398.22846 33969 0.44 0.5139
alpl 4.48057 4.31189 84211 1.08 0.3063
lva -163.00107 129.22589 124085 1.59 0.2160
gva 196.26305 84.74345 418315 5.36 0.0269
oid -24.38022 48.91350 19376 0.25 0.6215
fus -394.13999 80.32419 1877787 24.08 <.0001
p100 19.86216 18.57026 89219 1.14 0.2926

Bounds on condition number: 3.0797, 140.06


----------------------------------------------------------------------------

The above model is the best 8-variable model found.

No further improvement in R-Square is possible.

You might also like