Skip to content

FRM Part II · FRM Exam Part II · The Financial Stability Implications of Artificial Intelligence

A risk manager at an asset manager reviews an AI trading model whose training data covers only 2012-2021, a period of low inflation and low interest rates. The model performs well in backtests but degrades sharply when inflation spikes. Which model risk issue does this best illustrate?

The model has overfit to a narrow low-inflation, low-rate regime, so it generalises poorly when conditions change. Strong backtest results reflect the training period, not robustness, which is why validation should include out-of-sample and stressed scenarios.

  1. AOverfitting to a limited data regime, so the model generalises poorly to unseen conditionsCorrect
  2. BModel drift caused by improved data quality over time
  3. CLook-ahead bias caused by using future prices in the training sample
  4. DExcessive explainability that makes the model easy to game

Explanation

Training on a narrow regime produces a model that fits that regime but fails out of sample under different conditions, which is a limited-data or overfitting problem. Look-ahead bias would inflate backtests through use of future data, which the scenario does not describe. Data quality did not improve; it was unrepresentative.

Did you get it right without looking?

One question tells you little. A timed set on The Financial Stability Implications of Artificial Intelligence shows your real accuracy, how long you take and where you lose marks.

Practise The Financial Stability Implications of Artificial IntelligenceFree diagnostic mock

More The Financial Stability Implications of Artificial Intelligence questions