News

Scorista is expanding its capabilities by adding a new metric to its borrower creditworthiness assessment: bankruptcy prediction.

Our assessment now considers not only confirmed bankruptcy cases but also the likelihood that a borrower will face bankruptcy proceedings in the future.

How it works

We cross-referenced historical loan applications with subsequent changes in borrowers' credit histories. We then created a dataset comprising clients who later underwent bankruptcy proceedings and those who did not.

Based on this data, we developed several models with varying forecast horizons—ranging from 30 days to one year. The final forecast is generated using an ensemble of these models.

This scoring approach allows for the consideration of potential bankruptcy risk before it becomes evident in a borrower's credit history.

Why this matters for MFOs

A borrower's bankruptcy can directly impact loan repayment and the quality of the credit portfolio. The sooner an MFO receives a signal regarding potential risk, the more opportunities it has to factor that risk into its lending decisions.

This new metric can serve as an additional factor when assessing borrowers and fine-tuning an MFO's customized scoring strategy.

Read more about the model development methodology and the data used in this article by our mathematician.