half normal distribution fixed effect model


I want to use the fixed effect model for my panel/longitudinal data. The dependent variable is an absolute number, which in this case results in an half normal distribution (or folded normal distribution). Is this allowed or should I transform my data?

I am not a big fan of transforming the data because it is more difficult to really grasp the results. Any suggestions or examples on how to present research results with transformed data?

Thanks a lot for your help!!