The MDC Procedure

                                                 Conditional Logit Estimates

             Algorithm converged.


                                                      Model Fit Summary

                                         Dependent Variable                        y
                                         Number of Observations                 4353
                                         Number of Cases                      478830
                                         Log Likelihood                       -12358
                                         Maximum Absolute Gradient        5.16624E-7
                                         Number of Iterations                      7
                                         Optimization Method          Newton-Raphson
                                         AIC                                   24728
                                         Schwarz Criterion                     24766

                                                  Goodness-of-Fit Measures

                         Measure                       Value    Formula

                         Likelihood Ratio (R)          16206    2 * (LogL - LogL0)
                         Upper Bound of R (U)          40922    - 2 * LogL0
                         Aldrich-Nelson               0.7883    R / (R+N)
                         Cragg-Uhler 1                0.9758    1 - exp(-R/N)
                         Cragg-Uhler 2                0.9759    (1-exp(-R/N)) / (1-exp(-U/N))
                         Estrella                     0.9913    1 - (1-R/U)^(U/N)
                         Adjusted Estrella            0.9912    1 - ((LogL-K)/LogL0)^(-2/N*LogL0)
                         McFadden's LRI                0.396    R / U
                         Veall-Zimmermann             0.8721    (R * (U+N)) / (U * (R+N))

                         N = # of observations, K = # of regressors

                                                       The SAS System                       10:23 Wednesday, June 3, 2009 364

                                                      The MDC Procedure

                                                 Conditional Logit Estimates

                                                     Parameter Estimates

                                                                Standard                 Approx
                              Parameter     DF     Estimate        Error    t Value    Pr > |t|

                              tc             1      -0.0364     0.000536     -67.82     <.0001
                              drum           1       3.2083       0.2303      13.93     <.0001
                              trout          1       1.4674       0.0883      16.61     <.0001
                              drum2          1      -1.5574       0.1668      -9.34     <.0001
                              trout2         1      -0.2716       0.0209     -13.00     <.0001
                              lognsite       1       0.5926       0.0285      20.79     <.0001





                                                       The SAS System                       10:23 Wednesday, June 3, 2009 377

                                                      The MDC Procedure

                                                   Nested Logit Estimates

             Algorithm converged.


                                                      Model Fit Summary

                                         Dependent Variable                        y
                                         Number of Observations                 4353
                                         Number of Cases                      478830
                                         Log Likelihood                       -12351
                                         Maximum Absolute Gradient         0.0000763
                                         Number of Iterations                     12
                                         Optimization Method          Newton-Raphson
                                         AIC                                   24716
                                         Schwarz Criterion                     24760



                                                  Goodness-of-Fit Measures

                         Measure                       Value    Formula

                         Likelihood Ratio (R)          16221    2 * (LogL - LogL0)
                         Upper Bound of R (U)          40922    - 2 * LogL0
                         Aldrich-Nelson               0.7884    R / (R+N)
                         Cragg-Uhler 1                0.9759    1 - exp(-R/N)
                         Cragg-Uhler 2                 0.976    (1-exp(-R/N)) / (1-exp(-U/N))
                         Estrella                     0.9913    1 - (1-R/U)^(U/N)
                         Adjusted Estrella            0.9913    1 - ((LogL-K)/LogL0)^(-2/N*LogL0)
                         McFadden's LRI               0.3964    R / U
                         Veall-Zimmermann             0.8723    (R * (U+N)) / (U * (R+N))

                         N = # of observations, K = # of regressors

                                                       The SAS System                       10:23 Wednesday, June 3, 2009 382

                                                      The MDC Procedure

                                                   Nested Logit Estimates

                                                    Parameter Estimates

                                                                Standard                 Approx
                             Parameter      DF     Estimate        Error    t Value    Pr > |t|

                             tc_L1           1      -0.0366     0.000542     -67.46     <.0001
                             drum_L1         1       2.9722       0.2415      12.31     <.0001
                             trout_L1        1       1.6374       0.1077      15.20     <.0001
                             drum2_L1        1      -1.3797       0.1630      -8.47     <.0001
                             trout2_L1       1      -0.3111       0.0254     -12.27     <.0001
                             lognsite_L1     1       0.5956       0.0285      20.90     <.0001
                             INC_L2G1        1       0.5399       0.0942       5.73     <.0001