|
Decision
system construction, validation and regulatory review . |
|
To
contact us: |
|
Loss severity is correlated to geography, home price
appreciation, lender processes,
loan-to-value ratios, property type, loan size, and loan age. Logical assumptions, such as a
one-to-one relationship to LTV or
home price changes, are not proven
correct in actual
experience. Origination standards vary widely between lenders and loss severities vary
widely as a result. The President of
MDT authored a comprehensive
research study on loss severity in the December 1995 issue of The Journal of Fixed Income. One of the primary findings of the research was that the
identity of the lender was a significant variable in the prediction of
loss severity. |
|
Management Consulting Services |
|
Mortgage Decision Technologies, Inc. |
|
Loss severity should be integrated into any
sophisticated risk based pricing framework.
Credit Models rank order loans
by probability of becoming “Other Real Estate Owned” (OREO) .
While probability of becoming OREO is much more predictable than
severity, severity scoring is statistically predictive in estimation of loss
in basis points. A “loss in basis
points” estimate is the common denominator required to balance credit risk
with prepayment risk. |
