Dear all,
I have an unbalanced panel data and my DV is a proportion bonded between 0 and 1.
I have seen several answers to similar questions that refer to "Papke and Wooldridge 2008" (https://doi.org/10.1016/j.jeconom.2008.05.009) in that a working solution would be to use xtgee with family binomial and link logit:
But if i do understand the paper of Papke and Wooldrige correct, they only suggest it for balanced panel data. In a different post on statalist Mr. Wooldridge answers, that the command i have shown above is also applicable for unbalanced data (https://www.statalist.org/forums/for...ndent-variable.
So my question is: Can I use the shown command on unbalanced data and if so, are there any paper i could use as a source.
Thank you all and kind regards,
Peter
I have an unbalanced panel data and my DV is a proportion bonded between 0 and 1.
I have seen several answers to similar questions that refer to "Papke and Wooldridge 2008" (https://doi.org/10.1016/j.jeconom.2008.05.009) in that a working solution would be to use xtgee with family binomial and link logit:
Code:
xtgee DV IV, family(binomial) link(logit) corr(exchangeable) vce(robust)
So my question is: Can I use the shown command on unbalanced data and if so, are there any paper i could use as a source.
Thank you all and kind regards,
Peter
