MRM P&C - Model Drift & Performance Surveillance

Purpose

MRM P&C - Model Drift & Performance Surveillance is designed to continuously monitor input and output drift, as well as accuracy degradation, over time for insurance-related model monitoring activities.

Primary users

The primary user is not specified in the provided information. The agent is associated with the AQS team and owned by Ronan Davit.

Where it fits (process/stage/trigger)

This agent fits into ongoing model surveillance processes where historical runs, input data, and performance metrics are reviewed over time to identify drift or degradation.

Key capabilities / workflow

The agent continuously monitors historical runs, analyzes input and output drift, checks whether drift is detected, assesses performance metrics, evaluates whether accuracy has degraded, and generates surveillance results when relevant.

Inputs

The specified datasets are historical runs, input data, and performance metrics. Additional inputs are not specified in the provided information.

Outputs / Deliverables

The outputs are not specified in the provided information. Please refer to documentation for the expected deliverables generated by the agent.

Value

The agent supports model risk monitoring in insurance by helping identify model drift and accuracy degradation over time, enabling ongoing surveillance based on historical runs, input data, and performance metrics.