Modeling Mortgage Default as Loan-State Transitions
Summary
The discussion challenges a simple rule that declares a mortgage defaulted once a threshold share of payments is missed. A missed payment may be cured, and payment behavior can change as the borrower’s circumstances and macroeconomic conditions evolve. Instead of predicting a single binary outcome from payment-level predictions, the answer recommends modeling the loan’s progression through delinquency states.
A simplified whole-loan framework moves from current status to mild delinquency, then potentially to more serious delinquency, and finally to outcomes such as foreclosure, voluntary prepayment, or maturity. At each stage, the model estimates possible transitions, including returning to current status or curing the delinquency. Macro scenarios can be translated into loan-level predictors such as loan-to-value and borrower credit measures. The answer is conceptual rather than a validated model specification; it gives no transition probabilities, calibration evidence, or detailed treatment of servicing rules and recovery estimates.
Key ideas
- A missed payment alone does not establish that a mortgage will ultimately default.
- Credit models can represent a loan through delinquency stages and estimate transitions between them.
- Borrowers may cure delinquency and return to current status before foreclosure occurs.
- Foreclosure, voluntary prepayment, and maturity are distinct terminal outcomes to model.
- Macroeconomic scenarios can inform loan-level predictors used in transition estimates.
Tags
Full text
# Modeling mortgage loan defaults # Modeling mortgage loan defaults I have a machine learning model trained with a list of mortgage features that include macro variables where the field to predict (the label) is "Mortgage Defaulted" = 1 or 0 (Yes or No). Now, I need to determine if a mortgage will default in the future. For that, I use the same macro features where the model returns if a payment will be made or not. But if a single payment is not made, that doesn't mean that the mortgage will default, as the borrower may delay the payment. Moreover, if macro conditions improve, I may have a sequence of payments not made followed by a sequence of payments made. What is the best way to determine that the mortgage will default? My thinking is to have a percentage of payments NOT made, let's say 10%, and if the number of payments is greater than the percentage, then declare the mortgage as defaulted. Is this assumption valid? Any additional ideas will be welcome. ## Answer by Bond wiz (score 3, accepted) https://quant.stackexchange.com/a/46084 You should consider the stages of the default process instead of a binary "default", where there are various points the borrower is able to cure the loan. In a traditional credit model, the general process is to predict the state of the loan and then predict transitions between stages over the life of the loan. This is done by simulating macro variables (rates, HPI, employment) translated to individual loan predictors (current LTV, current FICO, HPA). Here is a simplified version of a whole loan model: - Current - from here can be prepaid (voluntary) or can become delinquent - Mildly Delinquent (first few months) - from here payments can be restarted and over time become current, the loan can be prepaid in full , or the borrower can move to #3 and become very delinquent. - Very Delinquent (5+ months or so) - from here, loan can be paid (back into mild dq bucket) or can continue to not pay and become an involuntary prepay (foreclosure). - Final stage - Either foreclosure after continued dq (need to estimate a recovery lag and amount from auction/REO sale), voluntary prepayment (refinance or relocation), or loan maturity. Agencies will buy back loans at 120 (conventional) or 180 (gnma) days but I assume from your question you are looking at whole loans where predicting all of the stages of loan and transitions between them are very important.
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.