|
|
Enron Mail |
---------------------- Forwarded by Zimin Lu/HOU/ECT on 03/21/2001 08:33 AM
--------------------------- From: Harry Arora/ENRON@enronXgate on 03/21/2001 07:26 AM To: Sanjay Gupta/ENRON@enronXgate, Steve Nat/ENRON@enronXgate cc: Zimin Lu/HOU/ECT@ECT Subject: Changes in option valuation in Enpower Sanjay Wanted to confirm the changes to the option valuations in the Enpower we discussed yesterday evening. 1. Currently the trader volatility inputs are the daily vol curve and the intra-monthly vol curve. The monthly options get marked directly to the monthly curve (plus the smile) and the monthlies get marked to a time blend of monthly and intra-month vol (plus the skew). We want to change the valuation (for the Eastern books) so that the dailies get marked to the intramonth curve (which we want to call the daily curve) and the monthly gets marked to the monthly curved. There will be not vol blending done by the Enpower system for the daily and monthly option valuations. We want to make this change very soon (by early next week) 2. Currently there exists one smile for every region, which is specified in terms of volatility additive for specified dollar difference from the underlying. Since different months in a region can trade in a large range ($35 - $150) - this cannot result in accurate skew for all terms. What we need is a system which has skew per month. We suggest, for the short term, the skew should apply only to the summer daily expiration options. We need to make this change by early next week. However, we need to start modifing the system so that for every region we can enter a grid which has a percentage scale and specifies the skew differently for each month. Research, has implemented this in our pricing model, and we would like this grid to be input into the valuation system. I am enclosing the pricing model (which we both discussed yesterday) for reference. This model is however, work under construction, so pls call Alex Huang for clarifications. 3. The vol input system is complex and confusing. I would very much be interested in moving to a direct database interface, which accomodates the skew inputs per region as in <2<. We should implement a UI which can input the dailies and monthlies at the moneys and the skew grid directly - so that we do not need to go through multiple iterations. I am very much interested in what we currently are releasing in Delphi and would love an early implementation for options. On all these issues, I am speaking for the East desk. I am going to touch base with West guys and see if they are on board with these changes. Thanks Harry
|