Need help about logitic regression with fixed effect and missing values

[So Rzzk]

Hi everyone!

I'm working on corruption's impact on the demand for bank credit. A study of discouraged borrowers in Africa with small firms.
However, I have some problem with the database.

In fact, I have merged several individual databases to create it with more or less the same variables and I end up with a lot of missing values.
In addition, I want to perform logistic regressions with fixed effects but I am stuck by missing values or having 10 times more 0 than 1. Therefore either I have r2000 errors, or stata is just turn round and round, turn but nothing happens or none of my variables are significant.
To solve the problem I carried out a multiple imputation but that does not change much.

I don't know what to do anymore, could you help me?

Thanking you for your help

[Felix Bittmann]

I am not sure but if your problem are the rare events, here is a summary of remedies you could try: https://www3.nd.edu/~rwilliam/stats3/rareevents.pdf

[So Rzzk]

Thank you very much!
However is it work with panel data? Because, in the database, I have one year per country and 3 countries at 3 different years.

[Rich Goldstein]

you should not be getting error message r(2000) unless you have lots of missing data or other data issues; please supply (1) a data example using -dataex- and pasted within CODE blocks (see the FAQ); (2) the exact command you types (again within CODE blocks) and (3) the exact message you are getting back from Stata; from what you say, you have about 9% of the data being "1" - this should, by itself, cause no problem