What can go wrong with the score test?

Please use this identifier to cite or link to this item: http://hdl.handle.net/1942/4016

Title:	What can go wrong with the score test?
Authors:	VERBEKE, Geert MOLENBERGHS, Geert
Issue Date:	2007
Publisher:	AMER STATISTICAL ASSOC
Source:	AMERICAN STATISTICIAN, 61(4). p. 289-290
Abstract:	The score test presents a number of issues. There are four dimensions of concern: (1) Is the parameter space constrained? (2) Is observed information used, or some form of expected information? (3) Are size and power computed under the null hypothesis, a correctly specified alternative, or a misspecified alternative? (4) Are computations asymptotic or small sample?
Notes:	Katholieke Univ Leuven, Ctr Biostat, Louvain, Belgium. Hasselt Univ, Ctr Stat, Diepenbeek, Belgium.VERBEKE, G, Katholieke Univ Leuven, Ctr Biostat, Louvain, Belgium.geert.verbeke@naed.kuleuven.be geert.molerlberghs@uhasselt.be
Keywords:	boundary condition; constrained parameter; expected information; likelihood ratio test; observed information; one-sided test; wald test;boundary condition; constrained parameter; expected information; likelihood ratio test; observed information; one-sided test; wald test
Document URI:	http://hdl.handle.net/1942/4016
ISSN:	0003-1305
e-ISSN:	1537-2731
DOI:	10.1198/000313007X243089
ISI #:	000250604700002
Rights:	(C) 2007 American Statistical Association
Category:	A1
Type:	Journal Contribution
Validations:	ecoom 2008
Appears in Collections:	Research publications

File	Description	Size	Format
27643925.pdf Restricted Access	Published version	399.45 kB	Adobe PDF	View/Open Request a copy

Check