Statalist


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

st: tobit or anything else?


From   Berthold Hoppe <berthold.hoppe@charite.de>
To   statalist@hsphsun2.harvard.edu
Subject   st: tobit or anything else?
Date   Sat, 12 Apr 2008 09:48:00 +0200

Dear Statalist users!

I try to analyze the relation between two continuous biological markers (y, x) in a study population, where -y- depends on -x-. Other data indicate that this relation is significantly influenced by a genotype z (two outcomes: 0, 1). The hypothesis is that the regression coefficients (y, x) differ between z=0 and z=1.
As -y- in censored (due to lower detection limit of the measurements) -tobit- instead of -reg- seems to be adequate.

Would the following analysis be correct to test my hypothesis?

xi3: tobit y x@z, ll
test x@z==0 = x@z==1

Does anyone know, how to handle this question.

Have a nice weekend,
Berthold

*
* 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