Template-Type: ReDIF-Article 1.0
Author-Name:  Sıdıka Başçı
Author-Workplace-Name: SESRIC
Author-Name:  Asad Zaman
Author-Email: asadzaman@alum.mit.edu
Author-Workplace-Name: International Islamic University of Islamabad
Author-Name:  Arzdar Kiracı
Author-Workplace-Name: Başkent University
Title:  Variance Estimates and Model Selection
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.
Classification-JEL: C13, C15, C22, C52
Keywords: Autoregressive Process, Lag Order Determination, Model Selection Criteria, Cross Validation
Journal: International Econometric Review
Pages: 57-72
Volume: 2
Issue: 2
Year: 2010
Month: September
File-URL: http://www.era.org.tr/makaleler/27040054.pdf
File-Format: Application/pdf
Handle: RePEc:erh:journl:v:2:y:2010:i:2:p:57-72