[Limdep Nlogit List] panel data set up

[Blane Lewis]

Hello Limdepers:

I am attempting to estimate one- and two-way linear and ordered choice fixed effects models.

I am explaining loan repayment arrears/default. I have loan data for 423 borrowers (i) across 23 time periods (t). In my data set, some borrowers have more than one loan in a given time period. That is, I have multiple observations for some indicator variables in some periods. 

Is this allowable? 

I get what appear to be reasonable results when I run the one- and two-way linear and one-way ordered choice models. But for the two-way ordered choice model I get an error message that says "time variable exceeds maximum period from pds". Perhaps the data set up is the problem?

Thanks in advance for any suggestions.

Blane