Hallo Statalist,

i am trying to find out the effect of the macro variables on the credit Rating in stress scenarios, or more precisely how many classes decreases the rating in stress scenarios (changing the macro variables).
the problem is I got my data from the company where i work and they include a huge percentage of Missing Values (about 85%).

The data include (CounterpartyNr , Rating Method, and Rating from 2005 to 2020) where the most customers have Missing information in there Rating that means i only have 4 or 5 customer out of 1380 with complete information (Rating) for the 16 years .

i tried to run a Linear Regression to my Panel the code xtreg and a logit Regression with xtlogit, but the results the i get from stata are not logical in term of the sign of the coefficient.

My Question is:
1. Do you think i can work with this data
2. is the logit regression the right one for me
3. can any other software solve my problem better than stata

The Dummy2 means the Rating goes 2 Rating classes down
The RealGDP and Exchange Rate should have a negative relationship with my dep. Variable, and all other independent Variables should have a positive relationship

Thanks