Re: st: A longitudinal model where the dependent variable is continuous between 0 and 1

[Najib Mozahem <najib.mozahem@gmail.com>]

Hi,
     Thanks for your answer. Since I can use the glm command to model the variable if it takes values from 0 to 1, that should mean that i can use the xtgee command with fam(binomial) and link(logit) to model the same variable if the dataset was longitudinal since generalized estimating equations are a generalization of generalized linear models?

Najib

On Aug 4, 2013, at 6:10 PM, Nick Cox wrote:

> Why tobit? See http://www.stata.com/statalist/archive/2013-08/msg00003.html
> and particularly the Baum reference there for the main point.
> Nick
> njcoxstata@gmail.com
> 
> 
> On 4 August 2013 15:58, Najib Mozahem <najib.mozahem@gmail.com> wrote:
>> Hi,
>>     My response variable can take on any value between 0 and 1. The records in my dataset belong to issues from newspapers and each newspaper has several issues, so I would like to take into account the clustering effect. I know about the betafit command with the cluster option, and I also know about the xtmetobit command a set the lower limit to 0 and the upper limit to 1. I would rather use the xtmetobit because I would like to model the random part as well, but I am not sure what to use to model the data.
>> 
>> Thanks
>> Najib
>> *
>> *   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/


*
*   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/