Two Paradoxes in Linear Regression Analysis

Regression is one of the favorite tools in applied statistics. However, misuse and misinterpretation of results from regression analysis are common in biomedical research. In this paper we use statistical theory and simulation studies to clarify some paradoxes around this popular statistical method....

Descripción completa

Detalles Bibliográficos
Autores principales: FENG, Ge, PENG, Jing, TU, Dongke, ZHENG, Julia Z., FENG, Changyong
Formato: Online Artículo Texto
Lenguaje:English
Publicado: Shanghai Municipal Bureau of Publishing 2016
Materias:
Biostatistics in Psychiatry (36)
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5434296/
https://www.ncbi.nlm.nih.gov/pubmed/28638214
http://dx.doi.org/10.11919/j.issn.1002-0829.216084
_version_ 1783237019944615936
author FENG, Ge
PENG, Jing
TU, Dongke
ZHENG, Julia Z.
FENG, Changyong
author_facet FENG, Ge
PENG, Jing
TU, Dongke
ZHENG, Julia Z.
FENG, Changyong
author_sort FENG, Ge
collection PubMed
description Regression is one of the favorite tools in applied statistics. However, misuse and misinterpretation of results from regression analysis are common in biomedical research. In this paper we use statistical theory and simulation studies to clarify some paradoxes around this popular statistical method. In particular, we show that a widely used model selection procedure employed in many publications in top medical journals is wrong. Formal procedures based on solid statistical theory should be used in model selection.
format Online
Article
Text
id pubmed-5434296
institution National Center for Biotechnology Information
language English
publishDate 2016
publisher Shanghai Municipal Bureau of Publishing
record_format MEDLINE/PubMed
spelling pubmed-54342962017-06-21 Two Paradoxes in Linear Regression Analysis FENG, Ge PENG, Jing TU, Dongke ZHENG, Julia Z. FENG, Changyong Shanghai Arch Psychiatry Biostatistics in Psychiatry (36) Regression is one of the favorite tools in applied statistics. However, misuse and misinterpretation of results from regression analysis are common in biomedical research. In this paper we use statistical theory and simulation studies to clarify some paradoxes around this popular statistical method. In particular, we show that a widely used model selection procedure employed in many publications in top medical journals is wrong. Formal procedures based on solid statistical theory should be used in model selection. Shanghai Municipal Bureau of Publishing 2016-12-25 2016-12-25 /pmc/articles/PMC5434296/ /pubmed/28638214 http://dx.doi.org/10.11919/j.issn.1002-0829.216084 Text en © Shanghai Municipal Bureau of Publishing http://creativecommons.org/licenses/by-nc/4.0/ This work is licensed under a Creative Commons Attribution-NonCommercial-Share Alike 4.0 Unported License. To view a copy of this license, visit http://creativecommons.org/licenses/by-nc-sa/4.0/
spellingShingle Biostatistics in Psychiatry (36)
FENG, Ge
PENG, Jing
TU, Dongke
ZHENG, Julia Z.
FENG, Changyong
Two Paradoxes in Linear Regression Analysis
title Two Paradoxes in Linear Regression Analysis
title_full Two Paradoxes in Linear Regression Analysis
title_fullStr Two Paradoxes in Linear Regression Analysis
title_full_unstemmed Two Paradoxes in Linear Regression Analysis
title_short Two Paradoxes in Linear Regression Analysis
title_sort two paradoxes in linear regression analysis
topic Biostatistics in Psychiatry (36)
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5434296/
https://www.ncbi.nlm.nih.gov/pubmed/28638214
http://dx.doi.org/10.11919/j.issn.1002-0829.216084
work_keys_str_mv AT fengge twoparadoxesinlinearregressionanalysis
AT pengjing twoparadoxesinlinearregressionanalysis
AT tudongke twoparadoxesinlinearregressionanalysis
AT zhengjuliaz twoparadoxesinlinearregressionanalysis
AT fengchangyong twoparadoxesinlinearregressionanalysis

Ejemplares similares