Variance Estimates and Model Selection

International Econometric Review. Vol. 2, No. 2, September 2010

17 Pages Posted: 10 Oct 2010  

Asad Zaman

Pakistan Institute of Development Economics

Sidika Basci

ESTIM Forecasting Center

Arzdar Kiraci

Başkent University

Date Written: October 8, 2010

Abstract

The large majority of the criteria for model selection are functions of the usual variance estimate for a regression model. The validity of the usual variance estimate depends on some assumptions, most critically the validity of the model being estimated. This is often violated in model selection contexts, where model search takes place over invalid models. A cross validated variance estimate is more robust to specification errors (see, for example, Efron, 1983). We consider the effects of replacing the usual variance estimate by a cross validated variance estimate, namely, the Prediction Sum of Squares (PRESS) in the functions of several model selection criteria. Such replacements improve the probability of finding the true model, at least in large samples.

Keywords: Autoregressive Process, Lag Order Determination, Model Selection Criteria, Cross Validation

JEL Classification: C13, C15, C22, C52

Suggested Citation

Zaman, Asad and Basci, Sidika and Kiraci, Arzdar, Variance Estimates and Model Selection (October 8, 2010). International Econometric Review. Vol. 2, No. 2, September 2010. Available at SSRN: https://ssrn.com/abstract=1689767

Asad Zaman (Contact Author)

Pakistan Institute of Development Economics ( email )

Quaid-i-Azam University Campus
P.O.Box 1091
Islamabad, Federal Capital 44000
Pakistan

Sidika Basci

ESTIM Forecasting Center ( email )

Sairler sok. 32/C
Gaziosmanpasa 06700
Turkey

Arzdar Kiraci

Başkent University ( email )

Bağlıca Kampüsü, Eskişehir Yolu 20. km.
Ankara
Turkey

Paper statistics

Downloads
45
Abstract Views
369