John,

We can approach this problem this way.  Basically you are asking how
the total variance (66%)2*143 are distributed among 113 days and the last?30 days.   Assuming volatility for the first period is simga1 and that in the ?last 30 ?days is sigma2, then??(66%)2*143=sigma12*113+sigma22*30

Futhermore, we can use Nov-00 implied volatility as a proxy to sigma1, then 
we can
calculate sigma2 which is the volatility for Dec-00 contract in the last 30 
days.

sigam2=sqrt((66%)2*143-sigma12*113)/30.

Make sense ?


Zimin







John Disturnal
07/09/2000 04:10 PM
To: Zimin Lu/HOU/ECT@ECT
cc:  
Subject: Vol rollup

Zimin, I am trying to understand what the near winter NG implied vols will 
look like as they approach expiration.  For example, is it possible to infer 
what the implied volatility of the Dec00 NG contract will look like with 30 
days to expiry given we know current vol (66%) and term to maturity ( 143 
days)?