No, in the binomial case, the data are sufficient. Eg.
input y freq
0 90
1 10
end
logit y [fw=freq]
The above is the same model as the glm case. Also, can get equivalent estimates with:
cii 100 10, wald
Yours,
Tim
-----Original Message-----
From: [email protected] [mailto:[email protected]] On Behalf Of Martin Weiss
Sent: 04 June 2009 17:45
To: [email protected]
Subject: st: Re: curious behaviour of -glm-
Which "other Stata commands" are you referring to? One of the most popular
ones exits with the same error message (which, BTW, is very clear: GET MORE
DATA!)
***
clear*
inp x
2
end
reg x
***
HTH
Martin
_______________
----- Original Message -----
From: "Mak, Timothy" <[email protected]>
To: <[email protected]>
Sent: Thursday, June 04, 2009 2:29 PM
Subject: st: curious behaviour of -glm-
> Hi list,
>
>
>
> I'm just curious why -glm, fam(bin) - doesn't seem to handle certain
> extreme scenarios in accordance with other Stata commands, like give you
> appropriate warning messages...
>
>
>
> Eg.
>
>
>
> . input r n
>
>
>
> r n
>
> 1. 10 100
>
> 2. end
>
>
>
> . glm r, fam(bin n)
>
> insufficient observations
>
> r(2001);
>
>
>
> The model should be estimable...
>
>
>
> Also:
>
>
>
>
>
> . input r n x
>
>
>
> r n x
>
> 1. 10 100 1
>
> 2. 0 100 0
>
> 3. end
>
>
>
> . glm r x, fam(bin n)
>
>
>
> Iteration 0: log likelihood = -2.2079271
>
> Iteration 1: log likelihood = -2.0261494
>
> Iteration 2: log likelihood = -2.0259832
>
> Iteration 3: log likelihood = -2.025974
>
> Iteration 4: log likelihood = -2.025974
>
>
>
> Generalized linear models No. of obs =
> 2
>
> Optimization : ML Residual df =
> 0
>
> Scale parameter =
> 1
>
> Deviance = 2.00000e-08 (1/df) Deviance =
> .
>
> Pearson = 1.00000e-08 (1/df) Pearson =
> .
>
>
>
> Variance function: V(u) = u*(1-u/n) [Binomial]
>
> Link function : g(u) = ln(u/(n-u)) [Logit]
>
>
>
> AIC =
> 4.025974
>
> Log likelihood = -2.025973987 BIC =
> 2.00e-08
>
>
>
> ------------------------------------------------------------------------------
>
> | OIM
>
> r | Coef. Std. Err. z P>|z| [95% Conf.
> Interval]
>
> -------------+----------------------------------------------------------------
>
> x | 23.87722 10000 0.00 0.998 -19575.76
> 19623.52
>
> _cons | -26.07444 10000 -0.00 0.998 -19625.71
> 19573.56
>
> ------------------------------------------------------------------------------
>
>
>
> The model runs as if there was convergence, but of course there can't
> be...
>
>
>
> Tim
>
>
>
> *
> * 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/
>
*
* 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/
*
* 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/
