You might want to consider the fractional logit model, proposed by Papke and Wooldridge:
Papke, L.E. & Wooldridge, J.M. (1996). Econometric methods for fractional response
variables with an application to 401(k) plan participation rates. Journal of Applied
Econometrics 11(6): 619–632.
See also: Papke, L.E. & Wooldridge, J.M. (2008). Panel data methods for fractional response
variables with an application to test pass rates. Journal of Econometrics 145: 121-133.
The Stata comand is:
glm y x1 x2...xk, family(binomial) link(logit) robust
There are other possibilities too. This persentation by Maarten Buis is very helpful:
http://www.dpc.de/fileadmin/templates/dpc-statistik/doc_pdf/add_info_stataUM/germany10_buis.pdf
Best,
Petya Alexandrova
>-------- Original Letter --------
>From: Geomina Turlea geomina@yahoo.fr
>Subject: st: I would like to restrict the coeficients in a regression so that the estimated values belong to a certain pre-defined interval. Is this possible?
>To: statalist@hsphsun2.harvard.edu
>Sent on: Tuesday, 2013, January 22 13:04:08 EET
> Hello everybody!
> I need to fill in missing data in a panel database of shares of ICT occupations in total occupations by country, industry and year. I am using a simple nested area unit level regression model. However, as the values of the depedent variable are shares, their predicted(estimated) value have to be restricted to the interval [0,1]. How can I do that using STATA? Thank you!
> Geomina
>
> *
> * For searches and help try:
> * http://www.stata.com/help.cgi?search
> * http://www.stata.com/support/faqs/resources/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/faqs/resources/statalist-faq/
* http://www.ats.ucla.edu/stat/stata/
