Modeling inverse relationship with gumbel copula

["Horace Tso" <Horace.Tso@pgn.com>]

Dear list,

I know it's late Friday afternoon (already night time on the east coast), I got 
stuck on this simple problem I couldn't resist going out to the list for 
advice. 

I want to model inverse relationship between two marginals with the gumbel 
copula. Say, x is lognormal and y is normal. Since gumbel's delta must be 
greater than one, how do I specify the equivalence of a negative correlation? 

Thanks in advance.

Horace W. Tso