Ordered Logit for panel data in Stata

[Nitin Jain]

Hi,

I have a dependent variable which is ordinal and the independent and control variables are continuous and categorical while the instruments are continuous.

I checked the Stata Menu: Statistics -> Endogenous Covariates -> Panel Data but am unable to find an option to perform ordinal logit on panel data with endogenous covariates.

Can someone please guide me how to perform it in Stata? I have noticed this type of model in research papers so I do think it is possible in Stata.

Any guidance is very much appreciated.

Thank you.

[Andrew Musau]

How many levels does the dependent variable have? Probably easiest to treat it as continuous and do linear IV regression.

help xtivreg

I have noticed this type of model in research papers so I do think it is possible in Stata.

Can you provide a reference?

[Joerg Luedicke (StataCorp)]

Nitin, you could consider using xteoprobit and fit an ordered probit model with endogenous regressors, see:

https://www.stata.com/manuals/ermeoprobit.pdf

[Nitin Jain]

Nitin, you could consider using xteoprobit and fit an ordered probit model with endogenous regressors, see:

https://www.stata.com/manuals/ermeoprobit.pdf

Thank you, Joerg.

[Nitin Jain]

How many levels does the dependent variable have? Probably easiest to treat it as continuous and do linear IV regression.

help xtivreg

Can you provide a reference?

Hello Andrew, There are 3 levels only. So, I am not sure if it would be fine to treat it as a continuous variable.

[Andrew Musau]

Hello Andrew, There are 3 levels only. So, I am not sure if it would be fine to treat it as a continuous variable.

No, that is too few.