Please use this identifier to cite or link to this item: https://hdl.handle.net/10419/31063 
Year of Publication: 
2006
Series/Report no.: 
Discussion Paper No. 478
Publisher: 
Ludwig-Maximilians-Universität München, Sonderforschungsbereich 386 - Statistische Analyse diskreter Strukturen, München
Abstract: 
A new method for testing linear restrictions in linear regression models is suggested. It allows to validate the linear restriction, up to a specified approximation error and with a specified error probability. The test relies on asymptotic normality of the test statistic, and therefore normality of the errors in the regression model is not required. In a simulation study the performance of the suggested method for model selection purposes, as compared to standard model selection criteria and the t-test, is examined. As an illustration we analyze the US college spending data from 1994.
Subjects: 
asymptotic normality
linear regression
model selection
model validation
Persistent Identifier of the first edition: 
Document Type: 
Working Paper

Files in This Item:
File
Size
309.12 kB





Items in EconStor are protected by copyright, with all rights reserved, unless otherwise indicated.