Automatic and Dynamic Models Backtesting

Client Needs & Objectives

Context:

  • Backtesting of banking models is time-consuming and extremely resource-intensive
  • Targeting the causes of poor performance allows to automate actions to take

Objectives:

Automation of model backtesting, calibration of alerts thresholds, identification of poor performances and actions to take

Scope:

  • Credit Risk
  • Asset management

Techniques:

  • Scoring
  • Classification
  • Logistic regression

Our Approach

  • A first layer of tests is implemented (Gini, Stability, Conservatism etc.).
  • According to the results, several layers of tests are built in order to identify the causes of model’s failures
  • Actions to take are automatically established
  • Thresholds are dynamically calibrated

Main Deliverables

Download