Dear list readers,
this is not an urgent question but rather a request for clarification.
I am a little confused about how Stata 7 deals with dropped variables in
regression and hope someone can explain it to me.
I am in a situation where I loop regressions over a certain condition and
save results in a variable, like this:
gen coeff=.
gen stdev=.
foreach d of numlist 1/20 {
regress y x if whatever==`d'
replace coeff=_b[x] if whatever==`d'
replace stdev=_se[x] if whatever==`d'
}
Now, in some cases Stata fails to compute the regression coefficient for x
and says "dropped" in the output. Apart from the question if this
regression makes any sense, I found that Stata saves 0 (zero) in _b[x],
and not missing value. I am confused by this behaviour, as I expected a
missing value or some error message saying "_b[x] is empty" or so.
Is there any explanation for this? At least theoretically there is a chance
for some regression coefficient being zero (or am I wrong here?).
Thanks a lot and best wishes,
Eva
____________________________________
Eva Poen
University of St. Gallen
Switzerland
*
* 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/
