[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
Re: st: linear probability model vs. probit/logit
At 03:04 PM 10/3/2006, Nishant Dass wrote:
Yes, but you should be getting a warning message about cases being
dropped if that is what is happening. In addition, you said you had
12,000 case for logit/probit and 10,000 for ols. Unless you had that
backwards, that would not be consistent with cases being dropped
because of perfect prediction, i.e. the N for logit/probit should be
smaller than the N for ols if perfect prediction is costing you cases.
Thanks for the link. I read it but I wonder - does perfect
prediction result in exclusion of those observations?
Is there anything else involved, like weights or clustering?
I checked again and my runs aren't different. I simply
replaced the -reg- with -logit- and re-run the command, and
get a different no. of obs. I am not sure how useful would
pasting my command be for you because there's really
nothing different between the two commands that I am
running (except the estimation method.)
Richard Williams, Notre Dame Dept of Sociology
OFFICE: (574)631-6668, (574)631-6463
WWW (personal): http://www.nd.edu/~rwilliam
WWW (department): http://www.nd.edu/~soc
* For searches and help try: