Re: st: Sample size for maximum likelihood estimates

[Maarten Buis <maartenlbuis@gmail.com>]

On Thu, Aug 11, 2011 at 5:02 PM, dk  wrote:
> What is the sample size needed for the MLE, if I have 20 explanatory
> variables and want to use Probit model. Can anyone send me the link
> for this problem.

To a large extend that depends on the distribution of your dependent
variable. In essence, you will get the most power (i.e. you can get by
with a smaller sample) when the probability of success is close to 50%
and you get less power (i.e. you need a larger sample) when the
probability of success gets closer to either 0% or 100%. The
distribution of your explanatory variables, the correlation between
these variables, and whether or not you want to add interaction
effects will also play a role. So I don't think there is a conclusive
answer to that question.

Hope this helps,
Maarten

--------------------------
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/