Effective Model Validation

Presenter Information/ Coauthors Information

Lance D. Cundy, Federal Reserve Bank of MinneapolisFollow

Presentation Type

Event

Student

No

Track

Finance/Insurance Application

Abstract

Model validation is a critical step in use of statistical and economic models. It is the set of processes and activities intended to verify that models are performing as expected, in line with their design objectives and business uses. In this session, we will discuss validation as it relates to financial and economic models. An effective validation framework should include three core elements: (i) evaluation of conceptual soundness, including developmental evidence, (ii) ongoing monitoring, including process verification and benchmarking, and (iii) outcomes analysis, including benchmarking. We will explore each of these elements and discuss statistical tools to support validation efforts.

Start Date

2-5-2019 1:00 PM

End Date

2-5-2019 1:50 PM

This document is currently not available here.

Share

Import Event to Google Calendar

COinS
 
Feb 5th, 1:00 PM Feb 5th, 1:50 PM

Effective Model Validation

Dakota Room 250 A/C

Model validation is a critical step in use of statistical and economic models. It is the set of processes and activities intended to verify that models are performing as expected, in line with their design objectives and business uses. In this session, we will discuss validation as it relates to financial and economic models. An effective validation framework should include three core elements: (i) evaluation of conceptual soundness, including developmental evidence, (ii) ongoing monitoring, including process verification and benchmarking, and (iii) outcomes analysis, including benchmarking. We will explore each of these elements and discuss statistical tools to support validation efforts.