June 18th, 2008, 11:55 pm
Hi Farshadb,(i) By simply truncating the replication, you will be missing the contribution of the Levy local time at the barrier, between now and maturity.To see that truncation is not sufficient, you can simply take a flat implied volatility surface: Although the fair strike of a range variance swap should be this vol value, the simple truncation method will give you a different result.(ii) Use the Meyer-Ito formula (cf. Protter's Stochastic Integration and Differential Equations book) to derive the replication formula.Regards,