Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down at the end of May, and its replacement, statalist.org is already up and running.


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

Re: st: Explaining apparent inconsistency in regression coefficients


From   Nick Cox <njcoxstata@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: Explaining apparent inconsistency in regression coefficients
Date   Fri, 8 Apr 2011 20:45:41 +0100

Your relation Y3 = Y1 + Y2 does not carry over to your models. Lots of
zeros on one Y could lever down the corresponding regressions.

You should be able to plot everything on a single graph to see what is
going on.

Nick

On Fri, Apr 8, 2011 at 8:23 PM, Hussain Samad <hsamad2000@yahoo.com> wrote:

> This is not exactly a STATA question except for the program is run in STATA. I
> am running a say OLS (not TOBIT) for three outputs:
>
> log(1+Y1)= alpha + beta*X + epsilon
> log(1+Y2)= alpha + beta*X + epsilon
> log(1+Y3)= alpha + beta*X + epsilon
>
> where Y1= farm income, Y2=nonfarm income, and Y3=total income=Y1+Y2, and X is a
> dummy intervention variable. Y1 or Y2 can be 0 but Y3 cannot. And these are the
> beta values I'm getting: 0.45 for Y1, 0.35 for Y2 and 0.21 for Y3. My
> question: why is the beta for Y3 is smaller than the other two? Shouldn't it be
> in between the other two betas? Any explanation will be highly appreciated.
>

*
*   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/


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