Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: Test the equality of regression coefficients


From   Simon Moore <simoncmoore@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   st: Test the equality of regression coefficients
Date   Wed, 04 Jul 2007 17:45:55 +0100

Dear Statalist,

I'm trying to test the test the equality of regression coefficients for two models. Looking at:

http://www.stata.com/support/faqs/stat/testing.html

I can see that the solution should be simple, but I'm stuck.

I'm modeling binary choice data across a series of trials per person to see how features of the choice environment effects choice for three groups. I started with the simplest model:-

tsset id trial
xi: xtlogit choice i.group*var1 var2, i(id) nolog re

Each group completed the task under two conditions and I want to compare
coefficients on the i.group*var1 interaction. So according to the above stata.com page I should append the data and then run something like

tsset id trial
xi3: xtlogit choice i.condition*i.group*var1 var2, i(id) nolog re

and then something like

test _IgroupXcondition_2 - _IgroupXcondition_3 = 0
local sign_ag = sign(_b[_IgroupXcondition_2] - _b[_IgroupXcondition_3])
display "H_0: group1 coef >= group2 coef. p-value = " normal(`sign_ag'*sqrt(r(chi2)))

But of course, I get

tsset id trial
repeated time values within panel

It seems like there should be a very simple way around this - but I'm stuck. So any hints would be greatly appreciated.

Regards
Simon


*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/




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