On 11 Nov 2000 13:46:30 -0800, [EMAIL PROTECTED] (Ick-Joong
Chung) wrote:
> Dear all,
>
> I have a question about two-sample problem. I am comparing coefficients
> of two samples (poor and non-poor) and would like to investigate whether
> the difference between two coefficients is statistically significant ('one
I'm already lost. Comparing two coefficients? Are you doing a
regression? You go on with detail that makes me think you are doing
two multiple regressions. But in that case, I don't know what
"two-sample t-statistic" is referred to, below --
> on one' level as well as 'overall' level). To compare coefficients from
> the two datasets in OLS settings, I can just use a two-sample t-statistic
> with a pooled variance estimate obtained from the models. I am wondering
> whether this can be applied to multinomial logistic regressions.
The Chow test is appropriate for either OLS or ML logistic; and it can
entail the use of terms computed as interactions...
> Alternatively, someone might suggest interaction terms. But unfortunately,
> it doesn't work for me because there are less power and sparsity issues
... and if you don't have power to draw conclusions, then you don't
have power. If the original regressions are barely 5% level, then the
difference between equations isn't likely to be shown. If you want to
focus a test on one or two variables, then you need to decide what
else should be present as covariates (not necessarily tested) and
focus on an intended test with one or two df.
> involved when I create as many interaction terms as predictors. And how
> can I do a F test for overall model comparison in multinomial logistic
> regressions?
>
> It's hard to find out ways of comparing coefficients in multinomial
> logistic regression settings. If you are aware of it, could you share it
> with me? It would be greatly appreciated.
--
Rich Ulrich, [EMAIL PROTECTED]
http://www.pitt.edu/~wpilib/index.html
=================================================================
Instructions for joining and leaving this list and remarks about
the problem of INAPPROPRIATE MESSAGES are available at
http://jse.stat.ncsu.edu/
=================================================================
