Why Prepayment Models Use State Transitions Instead of Amount Regression

Summary

The document asks why securitized-product prepayment models often represent borrower status with discrete delinquency states and estimate transition probabilities, rather than directly regressing the dollar amount prepaid in each period. It notes that observed prepayment amounts may be available in the data, making direct prediction seem like a simpler alternative to fitting separate logistic regressions for modeled transitions.

The text frames a modeling choice but does not provide an answer, evidence, or references. It therefore does not establish why transition models are preferred or whether direct amount regression is unsuitable. The question highlights a useful distinction between modeling borrower-state dynamics and predicting aggregate cash amounts, but any practical comparison would need to address the target definition, dependence over time, portfolio composition, and validation against the intended use.

Key ideas

• The document contrasts delinquency-state transition models with direct regression on prepayment amounts.
• A Markov framework models probabilities of moving between borrower states over time.
• The question observes that actual prepayment amounts may be available as regression targets.
• No explanation, empirical comparison, or literature references are supplied to resolve the modeling choice.

Tags

# Why isn't prepayment modelling done on actual prepaid amounts?


# Why isn't prepayment modelling done on actual prepaid amounts?












When modelling prepayments in Securitized products, why is it that the standard model involves a transition matrix (Markov Chain) framework, where the probabilities of transitioning between different states of delinquency are modelled? I have searched and have been unable to find any literature corresponding to models that are used to directly predict the actual prepayment amounts ($ amount) for each time period, given a dataset containing that information. As this information is directly available why isn't a regression fit directly to that data to predict the actual prepayment amount? I would like to understand why this seemingly easier method has been eschewed in favor of the Discrete-time Markov Chain framework for prepayment modelling (which requires logistic regressions for each transition that is modelled). If there is literature on such modelling, would be grateful if someone could provide the corresponding links.

Shown in full with attribution under the source's licence. Licence: CC BY-SA 4.0 (Stack Exchange)

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.