Hello, I've been studying a lot through statalist, but it's my first time posting a topic.
I have question on estimation method. I am so LOST.. on this. Any help will be appreciated.

Here is some characteristic of my data and variable.
- I have panel data t=9, n= about 35000.
- My dependent variable is heavily censored at 0 (about 40%) .
- Theoretically it is endogenous with one of my key independent variable A and B.
- No good IV available for A and B.

I think it will be best if I can use xttobit with good IV variable, but it is not possible.
So I have to think of way to address censored DV and endogeneity in panel data.

command 'craggit' seemed to address censored DV with endogeneity, but as far as I know, it is not suitable for panel data analysis.
(I guess craggit only can used as a pooled estimator?)

Any recommendation would be welcomed. Please help!