Prepayment is the single most important consideration in valuing
mortgage cash flows.
FIS Applied Analytics (formerly AFT) has produced a state-of-the-art behavioral
mortgage prepayment model that moves beyond traditional prepayment
models to provide a flexible trading-quality system.
The
model offers a powerful open architecture system that integrates
directly with your proprietary trading, portfolio management or
valuation system or can be used on a stand-alone basis.
This gives clients the flexibility to manage their portfolio
and make more informed valuation or trading decisions.
FIS Applied Analytics' prepayment model is stable. The model is based on borrowers'
reactions to changes in the environment that cause prepayments
(such as housing turnover or refinance). Because these reactions
change very slowly, FIS Applied Analytics' model is also slow to change.
Modeling Methodology
Prepayments are influenced by both borrower constraints and economic
incentives that change everyday.
Driving forces modeled by FIS Applied Analytics include: interest rate, structure
of burnout, details of origination information (points paid, extent
of premium coupon), age of loan, loan type, original loan to value,
loan balance and credit rating.
ARMS
Projecting ARM prepayments has been considered an extremely difficult
and, according to some, essentially impossible task. FIS Applied Analytics finds
that ARMs, although somewhat more volatile than fixed rate, are
quite tenable if one takes advantage of all of the information
available for an ARM.