Tom, Mark, Holden - below is how we are going to evaluate the UAMPS deal.
Please let me know if you have any questions.

Thanks,
Heather

 -----Original Message-----
From: 	Yang, Zhiyun  
Sent:	Monday, November 05, 2001 9:14 AM
To:	Dunton, Heather
Cc:	Huang, Alex; Lu, Zimin; Motley, Matt
Subject:	index option with index and delivery in different region

Hi Heather:

  Per our brief discussion last Friday, Alex asked me to put together a briefing on what research group thought was the easiest and reasonably accurate way to evaluate the index option when the index and delivery point were in different regions.

  Basically the research group think the option should be valued as a basis option for the whole period from the evaluation day to the expiration day, instead of for just one day ---- this is what we are doing in portcalc right now, except that the current evaluation is based on Craig vs. Rockie instead of Craig vs. PV.  The argument for that is that although the strike price won't be known till the last day, the whole price change process between today and then determines the final price.   I think Alex's example is quite eloquent:  say the basis difference between Craig and PV is zero today for 6/1/2002, it is very unlikely that the basis difference will still be zero on 5/31/2002.  If we evaluated as a one-day basis option whose basis price difference is one, the option value will be almost zero, and the option is underestimated.

  Put everything together, this is the suggested evaluation Research came up with:

  Underlying: forward hourly price for the expiration date in delivery region

  Strike: forward price for the delivery day in index region

  Time to expiration: time between evaluation date and expiration date.

  Alex and Zimin, please feel free to correct me if you find any error above.

  Thanks,

- Zhiyun