EViews.com

Hi,

I am using the panel date (42 cross-sections, 13 year time period) and want to correct/remove heteroskedasticity. I have tried to create log variables, but it does not work as many of the observations are zero. Any ideas how can I correct/remove heteroskedasticity?

Many Thanks,
F.

You could transform your variable say x to log(a+x) where a is a positive integer.