st: xi estimates table

["Jee-Yeon K. Lehmann" <jee@bu.edu>]

Hi.  When I tabulate the estimates results from a regression using xi or interaction expansion, the coefficients on certain variables appear as zeros or nulls.

Here's the code: 


xi: reg lwageNJ2 i.educat*black i.educat*hisp ageover if sex79==1 [aw=weightedu]
estimates store model1

xi: reg lwageNJ2 i.educat*black i.educat*hisp ageover mother father sib  cob res14 urban14 mom_us dad_us per_disad per_daily_attd per_dropout per_stud_asian per_stud_hisp per_stud_black lenroll lteachers llibrary lcouncel grad_teachers teachers_left teacher_salary if sex79==1 [aw=weightedu]
estimates store model2

xi: reg lwageNJ2 i.educat*black i.educat*hisp ageover grade80 if sex79==1&young==1 [aw=weightedu]
estimates store model3

xi: reg lwageNJ2 i.educat*black i.educat*hisp ageover grade80 mother father sib  cob res14 urban14 mom_us dad_us per_disad per_daily_attd per_dropout per_stud_asian per_stud_hisp per_stud_black lenroll lteachers llibrary lcouncel grad_teachers teachers_left teacher_salary if sex79==1&young==1 [aw=weightedu]
estimates store model4

estimates table model1 model2 model3 model4, stats(N) b(%9.4f) se(%6.4f)

...and I get this: 

  _Ieducat_2 |    0.1838                              0.0000  
             |    0.0303                              0.0000  
  _Ieducat_3 |    0.3976                              0.0000  
             |    0.0332                              0.0000  
  _Ieducat_4 |    0.6785                              0.0000  
             |    0.0345                              0.0000  
  _Ieducat_5 |    0.7891                              0.0000  
             |    0.0363                              0.0000  
       black |   -0.2686     -0.2349     -0.2673     -0.1670  
             |    0.0661      0.1411      0.1025      0.2207  
_IeduXbla~_2 |   -0.0261                             -0.0079  
             |    0.0730                              0.2276  
_IeduXbla~_3 |   -0.0576                             -0.1097  
             |    0.0827                              0.2652  
_IeduXbla~_4 |    0.0684                              0.1988  
             |    0.0963                              0.2696  
_IeduXbla~_5 |    0.1710                              0.2686  
             |    0.1214                              0.3553  


But when I simply look at the estimates form model2 alone, I get:

  _Ieducata2 |  .07657148  
  _Ieducata3 |  .25746476  
  _Ieducata4 |  .47787365  
  _Ieducata5 |  .52924125  
       black | -.23489743  
_IeduXbla~a2 |   .0029043  
_IeduXbla~a3 |  .03203529  
_IeduXbla~a4 |  .22297958  
_IeduXbla~a5 |  .13689507  


Thanks for your help.




*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/