eprintid: 20868
rev_number: 12
eprint_status: archive
userid: 2326
dir: disk0/00/02/08/68
datestamp: 2016-06-07 07:55:16
lastmod: 2016-06-30 08:54:34
status_changed: 2016-06-07 07:55:16
type: workingPaper
metadata_visibility: show
creators_name: Härdle, Wolfgang
creators_name: Mammen, Enno
creators_name: Müller, Marlene
title: Testing Parametric versus Semiparametric Modelling in Generalized Linear Models
subjects: ddc-510
divisions: i-110400
abstract: We consider a genralized partially linear model E(Y | X,T) = G{ X^T beta + m(T) } where G is a known function, beta is an unknown parameter vector, and m is an unknown function. The paper introduces a test statistic which allows to decide between a parametric and a semiparametric model:(i) m is linear, i.e. m(t) = t^T gamma for a parameter vector gamma,(ii) m is a smooth (nonlinear) function. Under linearity (i) it is shown that the test statistic is asymptotically normal. Moreover, it is proved that the bootstrap works asymptotically. Simulations suggest that (in small samples) bootstrap outperforms the calculation of critical values from the normal approximation. The practical performance of the test is shown in applications to data on East-West German migration and credit scoring.
date: 1998
id_scheme: DOI
id_number: 10.11588/heidok.00020868
schriftenreihe_cluster_id: sr-10a
schriftenreihe_order: 39
ppn_swb: 1657503143
own_urn: urn:nbn:de:bsz:16-heidok-208686
language: eng
bibsort: HARDLEWOLFTESTINGPAR1998
full_text_status: public
place_of_pub: Heidelberg
pages: 35
citation:   Härdle, Wolfgang ; Mammen, Enno ; Müller, Marlene  (1998) Testing Parametric versus Semiparametric Modelling in Generalized Linear Models.  [Working paper]     
document_url: https://archiv.ub.uni-heidelberg.de/volltextserver/20868/1/beitrag.39.pdf