The SAS System                       10:23 Wednesday, June 3, 2009 195

                                                      The MDC Procedure

                                                 Conditional Logit Estimates

             Algorithm converged.



                                                       The SAS System                       10:23 Wednesday, June 3, 2009 199

                                                      The MDC Procedure

                                                 Conditional Logit Estimates

                                                  Goodness-of-Fit Measures

                         Measure                       Value    Formula

                         Likelihood Ratio (R)         6051.8    2 * (LogL - LogL0)
                         Upper Bound of R (U)          14175    - 2 * LogL0
                         Aldrich-Nelson               0.7986    R / (R+N)
                         Cragg-Uhler 1                 0.981    1 - exp(-R/N)
                         Cragg-Uhler 2                0.9811    (1-exp(-R/N)) / (1-exp(-U/N))
                         Estrella                     0.9943    1 - (1-R/U)^(U/N)
                         Adjusted Estrella            0.9942    1 - ((LogL-K)/LogL0)^(-2/N*LogL0)
                         McFadden's LRI               0.4269    R / U
                         Veall-Zimmermann             0.8846    (R * (U+N)) / (U * (R+N))

                         N = # of observations, K = # of regressors

                                                       The SAS System                       10:23 Wednesday, June 3, 2009 200

                                                      The MDC Procedure

                                                 Conditional Logit Estimates

                                                     Parameter Estimates

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

                              tc             1      -0.0401     0.001061     -37.77     <.0001
                              small          1       0.1042       0.1003       1.04     0.2992
                              king           1       0.5249       0.9163       0.57     0.5668
                              spanish        1      -0.4916       0.1965      -2.50     0.0124
                              small2         1     0.003257       0.0199       0.16     0.8699
                              king2          1       0.8158       3.2697       0.25     0.8030
                              spanish2       1       0.0606       0.1184       0.51     0.6087
                              lognsite       1       0.6607       0.0463      14.26     <.0001

                                                       The SAS System                       10:23 Wednesday, June 3, 2009 201

                                                      The MDC Procedure

                                                   Nested Logit Estimates

             Algorithm converged.


                                                      Model Fit Summary

                                         Dependent Variable                        y
                                         Number of Observations                 1526
                                         Number of Cases                      158704
                                         Log Likelihood                        -4060
                                         Maximum Absolute Gradient         0.0001308
                                         Number of Iterations                     14
                                         Optimization Method          Newton-Raphson
                                         AIC                                    8137
                                         Schwarz Criterion                      8185



                                                       The SAS System                       10:23 Wednesday, June 3, 2009 205

                                                      The MDC Procedure

                                                   Nested Logit Estimates

                                                  Goodness-of-Fit Measures

                         Measure                       Value    Formula

                         Likelihood Ratio (R)         6055.5    2 * (LogL - LogL0)
                         Upper Bound of R (U)          14175    - 2 * LogL0
                         Aldrich-Nelson               0.7987    R / (R+N)
                         Cragg-Uhler 1                0.9811    1 - exp(-R/N)
                         Cragg-Uhler 2                0.9812    (1-exp(-R/N)) / (1-exp(-U/N))
                         Estrella                     0.9943    1 - (1-R/U)^(U/N)
                         Adjusted Estrella            0.9942    1 - ((LogL-K)/LogL0)^(-2/N*LogL0)
                         McFadden's LRI               0.4272    R / U
                         Veall-Zimmermann             0.8847    (R * (U+N)) / (U * (R+N))

                         N = # of observations, K = # of regressors

                                                       The SAS System                       10:23 Wednesday, June 3, 2009 206

                                                      The MDC Procedure

                                                   Nested Logit Estimates

                                                    Parameter Estimates

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

                             tc_L1           1      -0.0415     0.001309     -31.69     <.0001
                             small_L1        1       0.1263       0.1039       1.22     0.2243
                             king_L1         1       1.3712       1.0675       1.28     0.1990
                             spanish_L1      1      -0.3061       0.2275      -1.35     0.1785
                             small2_L1       1     0.002932       0.0202       0.15     0.8847
                             king2_L1        1      -1.1743       3.5690      -0.33     0.7421
                             spanish2_L1     1      -0.0193       0.1293      -0.15     0.8816
                             lognsite_L1     1       0.6639       0.0467      14.22     <.0001
                             INC_L2G1        1       0.8852       0.0563      15.73     <.0001