This looks like a valuable reference. From the abstract,
which is in the public domain, I note recommendation
of (among others) beta-distribution-based models for which -betafit-
on SSC may be of some use.
Note, however, the subtle but crucial notation point that
observations are on (0,1). The problem
which started this thread (see bottom) has observations
on [0,1). i.e. some values are 0. Does the
paper address this complication to anyone's knowledge?
Nick
n.j.cox@durham.ac.uk
Moran, John
> A good review of this problem (the authors used various
> packages, including
> Stata 7) is given in: R. Kieschnick and B. D. McCullough. Regression
> analysis of variates observed on (0,1): percentages, proportions and
> fractions. Statistical modelling 3:193-213, 2003.
TELHAJ Shqiponje
> The dependent variable I have is a proportion (percentage of 16 year
> olds enrolled in a particular subject) which is between 0 and 86
> percent. I am not sure about the linear form. My dependent
> variable is 0
> only in 3,980 cases out of 112,412 sample obs. Here a zero is a
> structural one, because the school does not offer history (which is
> choice subject).
>
> Would somebody suggest to me whether it would be better to perform a
> logit transformation, or estimate -glm- with family(gaussian) and
> link(logit). Any suggestion would be greatly appreciated!
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/