Naïve, Arima, Nonparametric, Transfer Function, and VAR Models: A Comparison of Forecasting Performance

29 Pages Posted: 12 Mar 2016

See all articles by Dimitrios D. Thomakos

Dimitrios D. Thomakos

University of Athens, Department of Business Administration

John Guerard

Independent

Date Written: June 1, 2002

Abstract

We examine the forecasting performance of parametric and nonparametric models based on a training-validation sample approach and the use of rolling short-term forecasts to compute root mean-squared errors,We find that the performance of these models is better than that of the naıve, no-change model. The use of bivariate models (like VAR and transfer functions) provides additional root mean-squared error reductions. In many cases the nonparametric models forecast as well or better than the parametric models. Our analysis suggests that (a) nonparametric models are attractive complements to parametric univariate models, and (b) simple VAR models should be considered before attempting to fit transfer function models.

Keywords: Forecasting, ARIMA

JEL Classification: C50; C53

Suggested Citation

Thomakos, Dimitrios D. and Guerard, John, Naïve, Arima, Nonparametric, Transfer Function, and VAR Models: A Comparison of Forecasting Performance (June 1, 2002). International Journal of Forecasting, Vol. 20, No. 1, 2004, Available at SSRN: https://ssrn.com/abstract=2745998

Dimitrios D. Thomakos (Contact Author)

University of Athens, Department of Business Administration ( email )

Athens
Greece

HOME PAGE: http://ba.uoa.gr/

John Guerard

Independent ( email )

Do you have negative results from your research you’d like to share?

Paper statistics

Downloads
73
Abstract Views
537
Rank
580,987
PlumX Metrics