CREDIT RISK
Credit risk assessment: follow portfolio risk over time
Credit risk assessment with TIMi measures and tracks the risk of a portfolio of commitments after the initial decision, and shows which factors explain each change.
Portfolio and file level
Risk estimated for segments of the portfolio or for individual files.
Follows the evolution
Updated as repayment history and borrower situations change.
Drift monitored
Modeler’s Drift Report shows when a model needs an update.
HOW IT WORKS
How credit risk assessment works.
Three steps, in one engine, on a standard server with no cluster.
1. Prepare the data
Anatella gathers exposure data, the repayment history since origination and signals of change in the borrower’s situation, then structures them into variables.
2. Build the models
Modeler generates models that estimate how risk evolves for segments of commitments or for individual files, taking into account changes since the original decision.
3. Assess and explain
Each commitment or portfolio segment gets an updated risk assessment, with the factors that explain its evolution.
HOW IT DIFFERS
Credit risk assessment versus credit scoring.
The two share the same modeling mechanics, but they answer different questions.
Granting versus following
Credit scoring evaluates an applicant at the time of the decision. Credit risk assessment covers commitments already running, or a whole portfolio, over time.
Anticipate deterioration
The aim is to detect a rise in risk before it turns into a default, for a segment or for a single file.
Prioritize actions
An updated score can rank the files to monitor or to contact first, before a payment incident occurs.
EXPLAINABILITY
A risk level says where. The factors say what changed.
An assessment alone gives a level, not what moved it. A white-box model identifies the factors behind a deterioration or an improvement for a segment or a file, which allows a targeted action instead of a global adjustment of the portfolio.
For risk teams
A change in risk can be documented factor by factor, which helps justify a monitoring action or a provision.
For regulatory work
The model supports your existing analysis with a continuous, explainable assessment. Specific regulatory obligations remain to be validated with your compliance teams.
For model owners
Modeler’s built-in Drift Report shows when a model loses accuracy, so you update it when needed instead of on a fixed calendar.
DATA
What data a credit risk model uses.
Anatella connects to more than 80 sources, so this data can be gathered without prior custom development.
Exposure and credit terms
Amounts, conditions and type of commitment.
Repayment history
Payments, delays and incidents since origination.
Signals of change
Income changes, payment incidents and sector context.
Portfolio segmentation
By type of commitment or borrower profile.
PROOF
Same models, applied to risk.
We do not yet have a published customer story dedicated to credit risk assessment. These deployments apply the same models to risk in banking and insurance.
98% accuracy on business bankruptcy within 6 months
Built from public balance sheets, with 24,000 variables including many year-over-year evolutions.
€250 million more could be invested safely
An individual-level model of cash-out probability showed that AXA could invest an additional €250 million without increasing liquidity risk.
FAQ
Common questions.
How often should a portfolio risk assessment be updated?
It depends on the volatility of the portfolio and on the economic context. Modeler includes a Drift Report that shows when a model has degraded, so you update it when needed instead of on an arbitrary schedule.
Can it help prioritize collection actions?
Yes. An updated risk score can rank the files to monitor or to contact first, before a payment incident occurs.
Do we need a long history?
A longer history improves precision, but a model can be built with a shorter one by relying on portfolio segments with comparable profiles.
Does it replace a regulatory risk analysis?
No. It supports the existing analysis with a continuous and explainable assessment, but specific regulatory obligations remain to be validated with your compliance teams.
RELATED USE CASES
Keep exploring.
SEE FOR YOURSELF
Test it on your own data.
Run TIMi on your own data with a free trial, or talk to us about your case.
