On the Selection of Forecasting Models

60 Pages Posted: 15 Mar 2003

See all articles by Atsushi Inoue

Atsushi Inoue

Southern Methodist University

Lutz Kilian

University of Michigan at Ann Arbor - Department of Economics; Centre for Economic Policy Research (CEPR)

Multiple version iconThere are 2 versions of this paper

Date Written: February 2003

Abstract

It is standard in applied work to select forecasting models by ranking candidate models by their PMSE in simulated out-of-sample (SOOS) forecasts. Alternatively, forecast models may be selected using information criteria (IC). We compare the asymptotic and finite-sample properties of these methods in terms of their ability to minimize the true out-of-sample PMSE, allowing for possible misspecification of the forecast models under consideration. We first study a covariance stationary environment. We show that under suitable conditions the IC method will be consistent for the best approximating model among the candidate models. In contrast, under standard assumptions the SOOS method will select overparameterized models with positive probability, resulting in excessive finite-sample PMSEs. We also show that in the presence of unmodelled structural change both methods will be inadmissible in the sense that they may select a model with strictly higher PMSE than the best approximating model among the candidate models.

Keywords: Model selection, Forecast accuracy, Structural change, Information criteria, Simulated out-of-sample method

JEL Classification: C22, C52, C53

Suggested Citation

Inoue, Atsushi and Kilian, Lutz, On the Selection of Forecasting Models (February 2003). ECB Working Paper No. 214. Available at SSRN: https://ssrn.com/abstract=387702

Atsushi Inoue

Southern Methodist University ( email )

Dallas, TX 75275
United States

Lutz Kilian (Contact Author)

University of Michigan at Ann Arbor - Department of Economics ( email )

611 Tappan Street
Ann Arbor, MI 48109-1220
United States
734-764-2320 (Phone)
734-764-2769 (Fax)

Centre for Economic Policy Research (CEPR)

London
United Kingdom

Register to save articles to
your library

Register

Paper statistics

Downloads
223
rank
124,975
Abstract Views
1,301
PlumX Metrics