Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: suest followed by hausman


From   "Schlesinger, Traci" <TSCHLESI@depaul.edu>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: suest followed by hausman
Date   Mon, 19 Oct 2009 11:15:10 -0500

dear stata users:

i am interested in finding out whether the affects of one variable (adj_red) on an individual's likelihood of being sentenced to incarceration (incar) varies depending on whether or not the individual is being sentenced in a state that uses sentencing guidelines (guidelines).  as such, i am using suest followed by a a hausman test.  however, currently my code works until i get to the test line, when i get the error: "equation [a_mean] not found / r(303);".  interestingly, if i run this same code using reg instead of logistic, the code works. 

what am i doing wrong?  the code is below.

thanks in advance for any feedback!

traci



#delimit ;

logistic  incar adj_red black latino age agesq murder rape robbery assault oviolent burglary theft MVT forgery fraud oproperty drugt odrug weapons driving chg1att  cjstatus priarr priorfta priconv pripris county year if guidelines == 1;

est store a;

logistic  incar adj_red black latino age agesq murder rape robbery assault oviolent burglary theft MVT forgery fraud oproperty drugt odrug weapons driving chg1att  cjstatus priarr priorfta priconv pripris county year if guidelines == 2;

est store b;

suest a b;

test [a_mean]adj_red=[b_mean]adj_red;

<<winmail.dat>>




© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index