st: Re: st: 答复: st: joint estimation of OLS and logit in seemingly unrelated regressions

[Maarten Buis <maartenlbuis@gmail.com>]

On Wed, May 23, 2012 at 4:04 PM, #WANG ZITIAN# wrote:
> Then, suppose my purpose is not to compare the statistical difference on the same variable across two regressions, but rather to adjust the correlated error terms across two regressions

The first thing I would do is think about what kind of information
from the data would be used for estimating such "corrections" and than
think about whether you find the resulting argument convincing. In all
situations where I have considered these types of models I have come
to the conclusion that there just wasn't enough information in the
data to justify such a method. Remember the following quote by John
Tukey (1986): "The combination of some data and an aching desire for
an answer does not ensure that a reasonable answer can be extracted
from a given body of data."

If you think your situation is different you can look at -cmp-, see:
-findit cmp-.

-- Maarten

John Tukey (1986), "Sunset salvo". The American Statistician 40(1):72-76.

--------------------------
Maarten L. Buis
Institut fuer Soziologie
Universitaet Tuebingen
Wilhelmstrasse 36
72074 Tuebingen
Germany


http://www.maartenbuis.nl
--------------------------
*
*   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/