Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: Count data models for panels with zero inflation?

From   "Wagner, Stefan" <[email protected]>
To   <[email protected]>
Subject   st: Count data models for panels with zero inflation?
Date   Fri, 28 Jan 2005 13:55:03 +0100

Dear Statalist,

I am analyzing a balanced panel dataset with the dependent variable of interest being a count variable. Additional to the observed count the upper bound of possible values is also known (i.e., I observe s successes in n trials for different groups i over several years t). 

Due to this data structure, I want to analyze the count variable conditional on the exogenous variables and conditional on the upper bound. 

I am not sure whether xtnbreg combined with the exposure-option (like in the code shown below) really solves my problem; the code I used is

xtnbreg depvar indepvars, exp(number of trials)

What makes the situation more complicated is that I got disproportionately many zeros in the data making a zero-inflated model more appropriate. I didn't find a clue within the STATA documentation whether any estimators for such a PANEL problem are implemented. Perhaps you can provide me with some guidance.

Any help is welcome!


FN:Wagner, Stefan
EMAIL;PREF;INTERNET:[email protected]

© Copyright 1996–2024 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index