Message-ID: <2933578.1075855467758.JavaMail.evans@thyme> Date: Thu, 6 Dec 2001 13:09:35 -0800 (PST) From: rabi.de@enron.com To: j.kaminski@enron.com Subject: FW: Margin Capital requirement -- back of the envelope simulation study Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-From: De, Rabi X-To: Kaminski, Vince J X-cc: X-bcc: X-Folder: \Vincent_Kaminski_Jan2002_1\Kaminski, Vince J\Inbox X-Origin: Kaminski-V X-FileName: vkamins (Non-Privileged).pst -----Original Message----- From: De, Rabi Sent: Thursday, December 06, 2001 1:39 PM To: Desai, Jayshree Cc: Tamarchenko, Tanya; Rajan, Karthik; Shanbhogue, Vasant Subject: Margin Capital requirement -- back of the envelope simulation study Jayshree, As requested you will find attached the spreadsheet containing results ( capital required and capital required normalized by VAR) for three levels of underlying volumes. The assumptions are numerous to list here -- so I list below only the major ones. 1. Underlyings in Natgas with a flat term structure -- $ 3.0 per MMBtu 2. A reasonable volatility term structure has been assumed 3. A representative set of 20 counterparties of various credit ratings selected on the basis of their MTM position on a random day this year. 4. The reported capital required is for a month and so is the VAR number 5. The volumetric exposure for each counterparty was assumed to be proportional to the MTM position on the same date as in item3. 6. We started with a premise of zero MTM, zero collateral at the starting time and zero threshold for margin calls all through. 7. The volume is distributed evenly between financial and physical contracts and across 60 forward months 8. No distinction has been made between physical and financial contracts 9. The margin calls are assumed to be a function of the exposure times a factor which varies from 0 to 1 as credit rating goes from AAA to D The results are therefore .... 1. the unnormalized results illustrate the trend of capital required with the volume multiplier but the absolute magnitude of the capital required is tied to the simulated VAR and hence to the underlying volume level assumed. If you have anymore questions please feel to call Karthik Rajan, Tanya Tamarchenko or myself. Regards, Rabi De