Everybody, 
we were talking for a while about using non-normal distributions in the 
Monte-Carlo simulations in our VAR model.
I put together some suggestion regarding this. The text is under 
O:\_Dropbox\Tanya\non_normal_logs.doc

Look through this 3 page document, and let me know what you think, please.


Tanya