Homoscedasticity: an overlooked critical assumption for linear regression

Linear regression is widely used in biomedical and psychosocial research. A critical assumption that is often overlooked is homoscedasticity. Unlike normality, the other assumption on data distribution, homoscedasticity is often taken for granted when fitting linear regression models. However, contr...

Full description

Saved in:
Bibliographic Details
Main Authors: Kun Yang, Justin Tu
Format: Article
Language:English
Published: BMJ Publishing Group 2019-10-01
Series:General Psychiatry
Online Access:https://gpsych.bmj.com/content/32/5/e100148.full
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850197516718440448
author Kun Yang
Justin Tu
author_facet Kun Yang
Justin Tu
author_sort Kun Yang
collection DOAJ
description Linear regression is widely used in biomedical and psychosocial research. A critical assumption that is often overlooked is homoscedasticity. Unlike normality, the other assumption on data distribution, homoscedasticity is often taken for granted when fitting linear regression models. However, contrary to popular belief, this assumption actually has a bigger impact on validity of linear regression results than normality. In this report, we use Monte Carlo simulation studies to investigate and compare their effects on validity of inference.
format Article
id doaj-art-9fc159ddfdaa439e931db18fac63d52e
institution OA Journals
issn 2517-729X
language English
publishDate 2019-10-01
publisher BMJ Publishing Group
record_format Article
series General Psychiatry
spelling doaj-art-9fc159ddfdaa439e931db18fac63d52e2025-08-20T02:13:07ZengBMJ Publishing GroupGeneral Psychiatry2517-729X2019-10-0132510.1136/gpsych-2019-100148Homoscedasticity: an overlooked critical assumption for linear regressionKun Yang0Justin Tu1Department of Evidence-based Medicine, Xuanwu Hospital Capital Medical University, Beijing, People’s Republic of China2 Department of Orthopedics, Emory Healthcare, Emory University, Atlanta, Georgia, USALinear regression is widely used in biomedical and psychosocial research. A critical assumption that is often overlooked is homoscedasticity. Unlike normality, the other assumption on data distribution, homoscedasticity is often taken for granted when fitting linear regression models. However, contrary to popular belief, this assumption actually has a bigger impact on validity of linear regression results than normality. In this report, we use Monte Carlo simulation studies to investigate and compare their effects on validity of inference.https://gpsych.bmj.com/content/32/5/e100148.full
spellingShingle Kun Yang
Justin Tu
Homoscedasticity: an overlooked critical assumption for linear regression
General Psychiatry
title Homoscedasticity: an overlooked critical assumption for linear regression
title_full Homoscedasticity: an overlooked critical assumption for linear regression
title_fullStr Homoscedasticity: an overlooked critical assumption for linear regression
title_full_unstemmed Homoscedasticity: an overlooked critical assumption for linear regression
title_short Homoscedasticity: an overlooked critical assumption for linear regression
title_sort homoscedasticity an overlooked critical assumption for linear regression
url https://gpsych.bmj.com/content/32/5/e100148.full
work_keys_str_mv AT kunyang homoscedasticityanoverlookedcriticalassumptionforlinearregression
AT justintu homoscedasticityanoverlookedcriticalassumptionforlinearregression