[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: regression with dependent variable ranging from 0 to 1

From   Maarten buis <>
Subject   Re: st: regression with dependent variable ranging from 0 to 1
Date   Tue, 30 Dec 2008 19:09:33 +0000 (GMT)

--- Andrea Rispoli <> wrote:
> Thank you. I tried this but I reach a message which says "Hessian has
> become unstable or asymmetric data not suitable for non-standard
> family-link combination".
> Maybe I should mention that my dependent variable is continuous.
> Is there anything I could do?

You did not say exactly what you typed, so it is hard to see what the
problem is, though the error message suggests that you mis-specified
the family for a fractional logit. The exact syntax for two of the
suggestions are illustrated below.

*------------------- begin example ---------------------------
use, clear

// betafit
ssc install betafit
betafit governing, mu(noleft minorityleft houseval popdens)

// fractional logit
glm governing noleft minorityleft houseval popdens, ///
    family(binomial) link(logit) robust
*------------------------ end example ------------------------
(For more on how to use examples I sent to the Statalist, see )

Hope this helps,

Maarten L. Buis
Department of Social Research Methodology
Vrije Universiteit Amsterdam
Boelelaan 1081
1081 HV Amsterdam
The Netherlands

visiting address:
Buitenveldertselaan 3 (Metropolitan), room N515

+31 20 5986715

*   For searches and help try:

© Copyright 1996–2017 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index