THE BUSINESS APPLICATION
Where and how to use it
Use it to understand roll rates, compare cohorts and explore future collections workload. Movement from early arrears to deeper arrears may call for a different response from a stable current portfolio, even when the aggregate overdue amount looks similar.
A lender sees a similar total overdue balance in two months. The transition view reveals that more accounts are now rolling into serious arrears while fewer return to current, prompting an operational review.
From evidence to a decision
- 01Starting stateCurrent or an arrears bucket
- 02Observed movementImprove · stay · deteriorate
- 03Transition matrixProbabilities over one interval
- 04Projected distributionExplore future arrears mix
DATA REQUIREMENTS
What records does it need?
These are the records your team would bring together for this analysis. The exact fields and history needed depend on your lending products, the question you want to answer and the period you want to assess.
| Record category | What it contains | Why the detail matters |
|---|---|---|
| Repeated account states | Account ID, observation date and a consistent arrears/default/closure state. | Comparable snapshots at a regular interval. |
| Balances and events | Exposure, cures, defaults, closures and restructures. | Distinguish account counts from balance-weighted movement. |
| Segmentation | Product, vintage and other permitted segment fields. | Each state and segment needs adequate observed support. |
Past loan outcomes help assess how well an estimate reflects your borrowers. For a new decision, use only the information available at that time; later repayments help you review the result afterwards.
Understand data readiness →WHAT YOU RECEIVE
The output
A transition matrix, state movements and conditional projections of account or balance distribution.
WHAT TO WATCH
The limitations
A state with no observations does not have a reliably estimated transition rate. Past roll rates may not persist after a change in policy, economic conditions or collections practice.
FOR RISK & ANALYTICAL SPECIALISTSHow the analysis works+
The modelling approach
Empirical or regularised transition probabilities are estimated between consistently defined states. Conditional models can distinguish segments where evidence permits. Multi-period projections depend on assumptions about how transition behaviour persists over time.
What your risk team should review
Rows sum to one, zero-support states are identified, state definitions are consistent and multi-period results reconcile.
The right approach depends on your portfolio and available history. Review the fit to your borrowers, the reliability of the estimates and the effect of missing information before using the result in a lending decision.