Re: st: tobit model in a percentage range

[Maarten Buis <maartenlbuis@gmail.com>]

> On Tue, Aug 16, 2011 at 11:57 AM, Cremades, Roger wrote:
>> I am hesitating about using Tobit in a model with a dep var in a
>>  percentage range with values in 0 or 100 <snip>

On Wed, Aug 17, 2011 at 9:12 AM, Tirthankar Chakravarty wrote:
> Percentage and proportion data are modelled by so-called fractional
> response regression models <snip>

A fractional logit model can be estimated in Stata using -glm-
together with the -link(logit) vce(robust)- option, see Baum (2008).
Other alternatives are -betafit-, -zoib-, -dirifit- and -fmlogit-, all
available from SSC. I compared all these different models in this talk
at the 2010 German Stata Users' meeting:
<http://ideas.repec.org/p/boc/dsug10/04.html>.

Hope this helps,
Maarten

Ps. All these models model proportions rather than percentages, so you
need to divide your dependent variable by 100.

Christopher F. Baum (2008) "Stata tip 63: Modeling proportions" The
Stata Journal, 8(2): 299-303.

--------------------------
Maarten L. Buis
Institut fuer Soziologie
Universitaet Tuebingen
Wilhelmstrasse 36
72074 Tuebingen
Germany


http://www.maartenbuis.nl
--------------------------
*
*   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/