Shipping example, 1 categorical and 2 cts. covs.

                                                        Prob
       Parameter    Level1    Estimate      StdErr     ChiSq

       Intercept               -9.6731      0.8698    <.0001
       type           A        -0.4060      0.2349    0.0840
       type           B        -0.9521      0.1953    <.0001
       type           C        -1.0386      0.3398    0.0022
       type           D        -0.6382      0.3014    0.0342
       type           E         0.0000      0.0000     . 
       built                    0.0422      0.0128    0.0010
       period                   0.0237      0.0081    0.0034
       Scale                    1.0000      0.0000     _ 
          Shipping example, 1 categorical and 2 cts. covs.
Criterion                     DF           Value         ValueDF
Deviance                      27         59.3746          2.1991
Scaled Deviance               27         59.3746          2.1991
Pearson Chi-Square            27         65.8510          2.4389
Scaled Pearson X2             27         65.8510          2.4389
Log Likelihood                 _        757.6734           _
Full Log Likelihood            _        -78.6205           _
AIC (smaller is better)        _        171.2410           _
AICC (smaller is better)       _        175.5487           _
BIC (smaller is better)        _        181.9256           _
              Shipping example, 2 categorical covs. 

                                                        Prob
       Parameter    Level1    Estimate      StdErr     ChiSq

       Intercept               -5.2564      0.2475    <.0001
       type           A        -0.3042      0.2358    0.1970
       type           B        -0.8629      0.1982    <.0001
       type           C        -0.9881      0.3393    0.0036
       type           D        -0.3746      0.3073    0.2229
       type           E         0.0000      0.0000     .
       built          60       -0.7083      0.2215    0.0014
       built          65        0.0452      0.2015    0.8225
       built          70        0.2570      0.1976    0.1933
       built          75        0.0000      0.0000     .
       Scale                    1.0000      0.0000     _
               Shipping example, 3 categorical covs. 

                                                        Prob
       Parameter    Level1    Estimate      StdErr     ChiSq

       Intercept               -5.2424      0.2473    <.0001
       type           A        -0.3256      0.2359    0.1675
       type           B        -0.8689      0.1984    <.0001
       type           C        -1.0130      0.3395    0.0028
       type           D        -0.4015      0.3071    0.1911
       type           E         0.0000      0.0000     .
       built          60       -0.4534      0.2332    0.0518
       built          65        0.2437      0.2088    0.2430
       built          70        0.3650      0.1995    0.0673
       built          75        0.0000      0.0000     .
       period         60       -0.3845      0.1183    0.0012
       period         75        0.0000      0.0000     .
       Scale                    1.0000      0.0000     _
            Result with interaction.  Note infinite ets. 

                                                            Prob
Parameter     Level1    Level2    Estimate      StdErr     ChiSq
Intercept                          -6.2953      1.0000    <.0001
type*built      A         60      -20.8915    58098.32    0.9997
type*built      A         65        0.5495      1.0690    0.6072
type*built      A         70        0.9835      1.0206    0.3352
type*built      A         75        0.9771      1.0445    0.3495
type*built      B         60       -0.5211      1.0073    0.6050
type*built      B         65        0.2056      1.0045    0.8378
type*built      B         70        0.4090      1.0089    0.6852
type*built      B         75        0.3154      1.0274    0.7589
type*built      C         60       -0.4680      1.2247    0.7023
type*built      C         65       -0.9889      1.4142    0.4844
type*built      C         70        0.4623      1.0607    0.6629
type*built      C         75        0.6821      1.4142    0.6296
type*built      D         60      -21.4876    57180.12    0.9997
type*built      D         65      -21.8581    59266.20    0.9997
type*built      D         70        1.5097      1.0377    0.1457
type*built      D         75        0.0555      1.1180    0.9604
type*built      E         60      -20.2045    84674.82    0.9998
type*built      E         65        1.8228      1.0351    0.0782
type*built      E         70        1.0214      1.0290    0.3209
type*built      E         75        0.0000      0.0000     .
Scale                               1.0000      0.0000     _
      Two parameterizations for Model ignoring race of victem 

                                                          Prob
  Parameter       Level1        Estimate      StdErr     ChiSq
  Intercept                       4.2556      0.0842    <.0001
  rdc*lc       Black Defend.     -2.1155      0.2567    <.0001
  rdc*lc       Black Defend.      0.0552      0.1175    0.6386
  rdc*lc       White Defend.     -2.0043      0.2444    <.0001
  rdc*lc       White Defend.      0.0000      0.0000     .
  Scale                           1.0000      0.0000     _
      Two parameterizations for Model ignoring race of victem 

                                                          Prob
  Parameter    Level1           Estimate      StdErr     ChiSq
  Intercept                       4.2556      0.0842    <.0001
  lc           Death             -2.0043      0.2444    <.0001
  lc           Live               0.0000      0.0000     .
  rdc          Black Defend.      0.0552      0.1175    0.6386
  rdc          White Defend.      0.0000      0.0000     .
  rdc*lc       Black Defend.     -0.1664      0.3539    0.6382
  rdc*lc       Black Defend.      0.0000      0.0000     .
  rdc*lc       White Defend.      0.0000      0.0000     .
  rdc*lc       White Defend.      0.0000      0.0000     .
  Scale                           1.0000      0.0000     _
        Simultaneously estimate all three common odds ratios
                                                        Prob
Parameter    Level1           Estimate      StdErr     ChiSq
Intercept                       4.8853      0.0868    <.0001
lc           Death             -1.9581      0.2451    <.0001
lc           Live               0.0000      0.0000     .
rdc          Black Defend.     -0.9403      0.1634    <.0001
rdc          White Defend.      0.0000      0.0000     .
rvc          Black Victim      -2.7249      0.3444    <.0001
rvc          White Victim       0.0000      0.0000     .
rdc*lc       Black Defend.      0.4402      0.4009    0.2722
rdc*lc       Black Defend.      0.0000      0.0000     .
rdc*lc       White Defend.      0.0000      0.0000     .
rdc*lc       White Defend.      0.0000      0.0000     .
rdc*rvc      Black Defend.      3.3580      0.3820    <.0001
rdc*rvc      Black Defend.      0.0000      0.0000     .
rdc*rvc      White Defend.      0.0000      0.0000     .
rdc*rvc      White Defend.      0.0000      0.0000     .
lc*rvc       Death             -1.3242      0.5193    0.0108
lc*rvc       Death              0.0000      0.0000     .  
lc*rvc       Live               0.0000      0.0000     .  
lc*rvc       Live               0.0000      0.0000     .  
Scale                           1.0000      0.0000     _  
               Traditional estimates for comparison.  

 Method              Value        LowerCL       UpperCL
 Mantel-Haenszel    1.5741         0.7096        3.4915
 Logit **           1.4545         0.6666        3.1735
         Poisson regression for Resp. Cancer Deaths
   Parameter    Level1    Estimate      StdErr     ChiSq
   Intercept               -5.8168      0.1911    <.0001
   period         1        -0.6926      0.2302    0.0026
   period         2        -0.1511      0.1803    0.4021
   period         3         0.0199      0.1542    0.8974
   period         4         0.0000      0.0000     .  
   hire           1         0.4828      0.1527    0.0016
   hire           2         0.0000      0.0000     .  
   ageg           1        -2.3088      0.2699    <.0001
   ageg           2        -0.9272      0.1835    <.0001
   ageg           3        -0.1396      0.1636    0.3933
   ageg           4         0.0000      0.0000     .  
   exp                      0.3174      0.0537    <.0001
   Scale                    1.0000      0.0000     _  
         Poisson regression for Circulatory Deaths 

                                                    Prob
   Parameter    Level1    Estimate      StdErr     ChiSq
   Intercept               -3.2464      0.0799    <.0001
   period         1        -0.1350      0.1007    0.1802
   period         2         0.0537      0.0813    0.5095
   period         3         0.0903      0.0705    0.2003
   period         4         0.0000      0.0000     .  
   hire           1        -0.0207      0.0714    0.7713
   hire           2         0.0000      0.0000     .  
   ageg           1        -2.6719      0.1039    <.0001
   ageg           2        -1.6185      0.0778    <.0001
   ageg           3        -0.8685      0.0710    <.0001
   ageg           4         0.0000      0.0000     .  
   exp                      0.0445      0.0295    0.1316
   Scale                    1.0000      0.0000     _  
       Logistic regression for proportional mortality  
                    The GENMOD Procedure

                         Model Information

             Data Set                      WORK.MONTANA
             Distribution                      Binomial
             Link Function                        Logit
             Response Variable (Events)             rcd
             Response Variable (Trials)           rcorc

              Number of Observations Read         114
              Number of Observations Used         107
              Number of Events                    276
              Number of Trials                   1599
              Number of Invalid Responses           7

                      Class Level Information
                   Class       Levels    Values
                   ageg             4    1 2 3 4 
                   period           4    1 2 3 4 
                   hire             2    1 2   


                          Response Profile
Ordered Value    Binary Outcome  Total Frequency
1                Event             276
2                Nonevent         1323

              Criteria For Assessing Goodness Of Fit
 Criterion                     DF           Value     Value/DF
 Deviance                      98        125.1902       1.2775
 Scaled Deviance               98        125.1902       1.2775
 Pearson Chi-Square            98        114.2425       1.1657
 Scaled Pearson X2             98        114.2425       1.1657
 Log Likelihood                         -712.0214
 Full Log Likelihood                    -153.9533
 AIC (smaller is better)                 325.9065
 AICC (smaller is better)                327.7622
 BIC (smaller is better)                 349.9620

Algorithm converged.  

         Analysis Of Maximum Likelihood Parameter Estimates
 
                        Standard       Wald 95%           Wald
Parameter   DF  Estimate  Error   Confidence Limits Chi-Square
Intercept    1   -2.4907 0.2040   -2.8905   -2.0909     149.09
period    1  1   -0.5554 0.2544   -1.0540   -0.0568       4.77
period    2  1   -0.2174 0.2015   -0.6123    0.1776       1.16
period    3  1   -0.1030 0.1724   -0.4410    0.2349       0.36
period    4  0    0.0000 0.0000    0.0000    0.0000        .  
hire      1  1    0.4479 0.1694    0.1159    0.7799       6.99
hire      2  0    0.0000 0.0000    0.0000    0.0000        .  
ageg      1  1    0.2911 0.2903   -0.2778    0.8600       1.01
ageg      2  1    0.6153 0.2000    0.2233    1.0073       9.46
ageg      3  1    0.6792 0.1796    0.3272    1.0312      14.30
ageg      4  0    0.0000 0.0000    0.0000    0.0000        .  
exp          1    0.2737 0.0629    0.1503    0.3970      18.91
Scale        0    1.0000 0.0000    1.0000    1.0000   

                        Analysis Of Maximum
                        Likelihood Parameter
                             Estimates
 
                      Parameter     Pr > ChiSq

                      Intercept         <.0001
                      period     1      0.0290
                      period     2      0.2807
                      period     3      0.5502
                      period     4       .
                      hire       1      0.0082
                      hire       2       .
                      ageg       1      0.3159
                      ageg       2      0.0021
                      ageg       3      0.0002
                      ageg       4       .
                      exp               <.0001
                      Scale

NOTE: The scale parameter was held fixed.
          Logistic regression for prostate date, ungrouped  

                                                        Prob
       Parameter    Level1    Estimate      StdErr     ChiSq

       Intercept                1.1486      0.3183    0.0003
       rx             1         0.4383      0.4849    0.3660
       rx             2         0.1671      0.4654    0.7196
       rx             3        -0.5890      0.4241    0.1649
       rx             4         0.0000      0.0000     .
       Scale                    1.0000      0.0000     _

         Logistic regression for prostate date, grouped   

                                                        Prob
       Parameter    Level1    Estimate      StdErr     ChiSq

       Intercept                1.1486      0.3183    0.0003
       rx             1         0.4383      0.4849    0.3660
       rx             2         0.1671      0.4654    0.7196
       rx             3        -0.5890      0.4241    0.1649
       rx             4         0.0000      0.0000     .
       Scale                    1.0000      0.0000     _
                         Probit regression 
                                                        Prob
       Parameter    Level1    Estimate      StdErr     ChiSq
       Intercept                0.7039      0.1868    0.0002
       rx             1         0.2510      0.2766    0.3642
       rx             2         0.0972      0.2705    0.7195
       rx             3        -0.3552      0.2545    0.1628
       rx             4         0.0000      0.0000     .    
       Scale                    1.0000      0.0000     _    
        Death penalty log. reg. w/o interactions to give MH.

PROC GENMOD is modeling the probability that lc='Death'. One 
way to change this to model the probability that lc='Live' is 
to specify the DESCENDING option in the PROC statement.

                              Standard     Wald 95%  Pr>
Parameter         DF Estimate   Error  Conf.  Limits ChiSq

Intercept          1  -1.9581  0.2451 -2.4384 -1.4778 <.0001
rvc  Black Victim  1  -1.3242  0.5193 -2.3421 -0.3063 0.0108
rvc  White Victim  0   0.0000  0.0000  0.0000  0.0000 .
rdc  Black Defend. 1   0.4402  0.4009 -0.3455  1.2260 0.2722
rdc  White Defend. 0   0.0000  0.0000  0.0000  0.0000 .
Scale              0   1.0000  0.0000  1.0000  1.0000 .

                Death penalty data with interactions
               Criteria For Assessing Goodness Of Fit
        Analysis Of Maximum Likelihood Parameter Estimates
                                       Standard Pr > ChiSq
Parameter              DF   Estimate      Error 
Intercept                1    -1.9384     0.2454 <.0001
rvc     B. V.            1   -22.4270     0.5358 <.0001
rvc     W. V.            0     0.0000     0.0000
rdc     B. D.            1     0.3850     0.4127 0.3509
rdc     W. D.            0     0.0000     0.0000
rdc*rvc B. D.   B. V.    0    21.1974     0.0000
rdc*rvc B. D.   W. V.    0     0.0000     0.0000
rdc*rvc W. D.   B. V.    0     0.0000     0.0000
rdc*rvc W. D.   W. V.    0     0.0000     0.0000
Scale                    0     1.0000     0.0000
NOTE: The scale parameter was held fixed.

                 LR Statistics For Type 3 Analysis
                                      Chi-
            Source           DF     Square    Pr > ChiSq

            rvc               1       5.36        0.0206
            rdc               1       1.40        0.2361
            rdc*rvc           1       0.70        0.4025