The Performance of Random Coefficient Regression in Accounting for Residual Confounding
Summary Greenland (2000, Biometrics 56, 915–921) describes the use of random coefficient regression to adjust for residual confounding in a particular setting. We examine this setting further, giving theoretical and empirical results concerning the frequentist and Bayesian performance of random coef...
Published in: | Biometrics |
---|---|
Main Authors: | , |
Format: | Article in Journal/Newspaper |
Language: | English |
Published: |
Wiley
2006
|
Subjects: | |
Online Access: | http://dx.doi.org/10.1111/j.1541-0420.2005.00510.x https://api.wiley.com/onlinelibrary/tdm/v1/articles/10.1111%2Fj.1541-0420.2005.00510.x https://onlinelibrary.wiley.com/doi/pdf/10.1111/j.1541-0420.2005.00510.x |
id |
crwiley:10.1111/j.1541-0420.2005.00510.x |
---|---|
record_format |
openpolar |
spelling |
crwiley:10.1111/j.1541-0420.2005.00510.x 2023-12-03T10:23:34+01:00 The Performance of Random Coefficient Regression in Accounting for Residual Confounding Gustafson, Paul Greenland, Sander 2006 http://dx.doi.org/10.1111/j.1541-0420.2005.00510.x https://api.wiley.com/onlinelibrary/tdm/v1/articles/10.1111%2Fj.1541-0420.2005.00510.x https://onlinelibrary.wiley.com/doi/pdf/10.1111/j.1541-0420.2005.00510.x en eng Wiley http://onlinelibrary.wiley.com/termsAndConditions#vor Biometrics volume 62, issue 3, page 760-768 ISSN 0006-341X 1541-0420 Applied Mathematics General Agricultural and Biological Sciences General Immunology and Microbiology General Biochemistry, Genetics and Molecular Biology General Medicine Statistics and Probability journal-article 2006 crwiley https://doi.org/10.1111/j.1541-0420.2005.00510.x 2023-11-09T13:45:45Z Summary Greenland (2000, Biometrics 56, 915–921) describes the use of random coefficient regression to adjust for residual confounding in a particular setting. We examine this setting further, giving theoretical and empirical results concerning the frequentist and Bayesian performance of random coefficient regression. Particularly, we compare estimators based on this adjustment for residual confounding to estimators based on the assumption of no residual confounding. This devolves to comparing an estimator from a nonidentified but more realistic model to an estimator from a less realistic but identified model. The approach described by Gustafson (2005, Statistical Science 20, 111–140) is used to quantify the performance of a Bayesian estimator arising from a nonidentified model. From both theoretical calculations and simulations we find support for the idea that superior performance can be obtained by replacing unrealistic identifying constraints with priors that allow modest departures from those constraints. In terms of point‐estimator bias this superiority arises when the extent of residual confounding is substantial, but the advantage is much broader in terms of interval estimation. The benefit from modeling residual confounding is maintained when the prior distributions employed only roughly correspond to reality, for the standard identifying constraints are equivalent to priors that typically correspond much worse. Article in Journal/Newspaper Greenland Wiley Online Library (via Crossref) Greenland Biometrics 62 3 760 768 |
institution |
Open Polar |
collection |
Wiley Online Library (via Crossref) |
op_collection_id |
crwiley |
language |
English |
topic |
Applied Mathematics General Agricultural and Biological Sciences General Immunology and Microbiology General Biochemistry, Genetics and Molecular Biology General Medicine Statistics and Probability |
spellingShingle |
Applied Mathematics General Agricultural and Biological Sciences General Immunology and Microbiology General Biochemistry, Genetics and Molecular Biology General Medicine Statistics and Probability Gustafson, Paul Greenland, Sander The Performance of Random Coefficient Regression in Accounting for Residual Confounding |
topic_facet |
Applied Mathematics General Agricultural and Biological Sciences General Immunology and Microbiology General Biochemistry, Genetics and Molecular Biology General Medicine Statistics and Probability |
description |
Summary Greenland (2000, Biometrics 56, 915–921) describes the use of random coefficient regression to adjust for residual confounding in a particular setting. We examine this setting further, giving theoretical and empirical results concerning the frequentist and Bayesian performance of random coefficient regression. Particularly, we compare estimators based on this adjustment for residual confounding to estimators based on the assumption of no residual confounding. This devolves to comparing an estimator from a nonidentified but more realistic model to an estimator from a less realistic but identified model. The approach described by Gustafson (2005, Statistical Science 20, 111–140) is used to quantify the performance of a Bayesian estimator arising from a nonidentified model. From both theoretical calculations and simulations we find support for the idea that superior performance can be obtained by replacing unrealistic identifying constraints with priors that allow modest departures from those constraints. In terms of point‐estimator bias this superiority arises when the extent of residual confounding is substantial, but the advantage is much broader in terms of interval estimation. The benefit from modeling residual confounding is maintained when the prior distributions employed only roughly correspond to reality, for the standard identifying constraints are equivalent to priors that typically correspond much worse. |
format |
Article in Journal/Newspaper |
author |
Gustafson, Paul Greenland, Sander |
author_facet |
Gustafson, Paul Greenland, Sander |
author_sort |
Gustafson, Paul |
title |
The Performance of Random Coefficient Regression in Accounting for Residual Confounding |
title_short |
The Performance of Random Coefficient Regression in Accounting for Residual Confounding |
title_full |
The Performance of Random Coefficient Regression in Accounting for Residual Confounding |
title_fullStr |
The Performance of Random Coefficient Regression in Accounting for Residual Confounding |
title_full_unstemmed |
The Performance of Random Coefficient Regression in Accounting for Residual Confounding |
title_sort |
performance of random coefficient regression in accounting for residual confounding |
publisher |
Wiley |
publishDate |
2006 |
url |
http://dx.doi.org/10.1111/j.1541-0420.2005.00510.x https://api.wiley.com/onlinelibrary/tdm/v1/articles/10.1111%2Fj.1541-0420.2005.00510.x https://onlinelibrary.wiley.com/doi/pdf/10.1111/j.1541-0420.2005.00510.x |
geographic |
Greenland |
geographic_facet |
Greenland |
genre |
Greenland |
genre_facet |
Greenland |
op_source |
Biometrics volume 62, issue 3, page 760-768 ISSN 0006-341X 1541-0420 |
op_rights |
http://onlinelibrary.wiley.com/termsAndConditions#vor |
op_doi |
https://doi.org/10.1111/j.1541-0420.2005.00510.x |
container_title |
Biometrics |
container_volume |
62 |
container_issue |
3 |
container_start_page |
760 |
op_container_end_page |
768 |
_version_ |
1784271750325862400 |