ADDITIVE MODELS WITH PREDICTORS SUBJECT TO MEASUREMENT ERROR

Summary This paper develops a likelihood‐based method for fitting additive models in the presence of measurement error. It formulates the additive model using the linear mixed model representation of penalized splines. In the presence of a structural measurement error model, the resulting likelihood...

Full description

Saved in:
Bibliographic Details
Published inAustralian & New Zealand journal of statistics Vol. 47; no. 2; pp. 193 - 202
Main Authors Ganguli, Bhaswati, Staudenmayer, John, Wand, M.P.
Format Journal Article
LanguageEnglish
Published Oxford, UK Blackwell Publishing Ltd 01.06.2005
Subjects
Metropolis-Hastings
mixed models
Monte Carlo expectation maximization
nested EM
penalized splines
restricted maximum likelihood
Online AccessGet full text
ISSN1369-1473
1467-842X
DOI10.1111/j.1467-842X.2005.00383.x

Cover

More Information
Summary:Summary This paper develops a likelihood‐based method for fitting additive models in the presence of measurement error. It formulates the additive model using the linear mixed model representation of penalized splines. In the presence of a structural measurement error model, the resulting likelihood involves intractable integrals, and a Monte Carlo expectation maximization strategy is developed for obtaining estimates. The method's performance is illustrated with a simulation study.
Bibliography:ArticleID:ANZS383
istex:5D0FD9CE7ECE3AAC174062852DC9B42CE66F1728
ark:/67375/WNG-47GTQGQL-V
This work was partially supported by NIH grant T32 ES07142‐18 and NSF‐DMS 0306227. The authors thank the editor, an associate editor, and an anonymous referee for helpful suggestions and comments.
Dept of Mathematics and Statistics, Lederle Graduate Research Tower, University of Massachusetts, Amherst MA, 01003‐9305, USA. e‐mail
jstauden@math.umass.edu
Acknowledgments.
Indian Institute of Management, Calcutta, India.
Dept of Statistics, The University of New South Wales, Sydney NSW 2052, Australia.
ISSN:1369-1473
1467-842X
DOI:10.1111/j.1467-842X.2005.00383.x