Message-ID: <10060208.1075856752489.JavaMail.evans@thyme>
Date: Tue, 4 Apr 2000 08:36:00 -0700 (PDT)
From: martina.angelova@enron.com
To: zimin.lu@enron.com
Subject: UK RPI model
Cc: vince.kaminski@enron.com, anjam.ahmad@enron.com, trena.mcfarland@enron.com
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Bcc: vince.kaminski@enron.com, anjam.ahmad@enron.com, trena.mcfarland@enron.com
X-From: Martina Angelova
X-To: Zimin Lu
X-cc: Vince J Kaminski, Anjam Ahmad, Trena McFarland
X-bcc: 
X-Folder: \Vincent_Kaminski_Jun2001_6\Notes Folders\All documents
X-Origin: Kaminski-V
X-FileName: vkamins.nsf

Hi Zimin!

Please find attached for your  review the UK RPI model, derived by 
bootstrapping RPI swaps. 

It's a very simple model and here are its specifics:

SWAP STRUCTURE

Payment: semi/semi Act/365F
>
> YOYUKRPI = (UKRPI(p-2)/UKRPI(p-14) - 1)/2
> p = payment month
>

The first payment is the latest known historical RPI, February 2000, 2.32%. 

ASSUMPTIONS

* Constant cashflows between the quoted years (as opposed to interpolating 
swaps which distorts the curve a lot). This explains the atrocious look of 
the "raw" curve. It is then smoothed with a macro, which Anjam wrote.

* Mid point of the swaps is used for deriving the curve;

* Discount rate is LIBOR and I solve for the coupon rate, which is the RPI 
YOY rate;



* The above is solved separately for each quoted period (e.g. 2 yrs, 5 yrs) 
and RPI rates are determined for the incremental portion.

By forecasting RPI in the above method we are able to lock in and deliver the 
forecasted levels.

Looking forward to your comments and seeing you in London!

Best regards,

Martina
x34327
