Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: st: Small standard errors of the parameters in logit models
From
Nick Cox <[email protected]>
To
[email protected]
Subject
Re: st: Small standard errors of the parameters in logit models
Date
Thu, 15 Dec 2011 11:15:03 +0000
There is no evidence here that you need any extra adjustment for
sample size beyond what is provided automatically and no indication
of what it should be even if your view is correct.
Statistical significance testing historically had one leading role, to
stop researchers making fools of themselves by over-interpreting
apparent effects from very small samples. And that is still of some
importance. However, with very large sample sizes it is no surprise
that almost anything will be significant, meaning usually definitely
not zero, and the question becomes one of interpreting the magnitude
of effects. As you report that the effects are small, that is also
consistent with what is typical for such datasets.
Nick
2011/12/15 Kai Huang <[email protected]>:
> Dear all,
>
> I have run a logit model on the employment probability in the labour force using pooled UK LFS data over various years. The estimated parameters are small in magnitude but highly statistically significant. I doubt that it is due to the large sample size rather than proper specification of the model. Does anyone know whether there are any STATA packages that estimate a limited dependent variable model with adjustment of standard errors to large sample size?
>
*
* 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/