Simple Bayesian testing of scientific expectations in linear regression models

Joris Mulder, Anton Olsson Collentine

Research output: Contribution to journal › Article › Scientific › peer-review

Abstract

Scientific theories can often be formulated using equality and order constraints on the relative effects in a linear regression model. For example, it may be expected that the effect of the first predictor is larger than the effect of the second predictor, and the second predictor is expected to be larger than the third predictor. The goal is then to test such expectations against competing scientific expectations or theories. In this paper a simple default Bayes factor test is proposed for testing multiple hypotheses with equality and order constraints on the effects of interest. The proposed testing criterion can be computed without requiring external prior information about the expected effects before observing the data. The method is implemented in R-package called `{\tt lmhyp}' which is freely downloadable and ready to use. The usability of the method and software is illustrated using empirical applications from the social and behavioral sciences.
Original languageEnglish
JournalBehavior Research Methods
Publication statusAccepted/In press - 2019

Fingerprint

Linear Models
Testing
Linear Regression
Predictors
Equality

Keywords

  • Bayes factors
  • regression modeling
  • Bayesian hypothesis testing
  • equality and order constraints
  • Psychology

Cite this

Mulder, Joris ; Olsson Collentine, Anton. / Simple Bayesian testing of scientific expectations in linear regression models. In: Behavior Research Methods. 2019.
@article{24432d78ce714cbcb25a0497fc2ccb86,
title = "Simple Bayesian testing of scientific expectations in linear regression models",
abstract = "Scientific theories can often be formulated using equality and order constraints on the relative effects in a linear regression model. For example, it may be expected that the effect of the first predictor is larger than the effect of the second predictor, and the second predictor is expected to be larger than the third predictor. The goal is then to test such expectations against competing scientific expectations or theories. In this paper a simple default Bayes factor test is proposed for testing multiple hypotheses with equality and order constraints on the effects of interest. The proposed testing criterion can be computed without requiring external prior information about the expected effects before observing the data. The method is implemented in R-package called `{\tt lmhyp}' which is freely downloadable and ready to use. The usability of the method and software is illustrated using empirical applications from the social and behavioral sciences.",
keywords = "Bayes factors, regression modeling, Bayesian hypothesis testing, equality and order constraints, Psychology",
author = "Joris Mulder and {Olsson Collentine}, Anton",
year = "2019",
language = "English",
journal = "Behavior Research Methods",
issn = "1554-351X",
publisher = "Springer",

}

Mulder, J & Olsson Collentine, A 2019, 'Simple Bayesian testing of scientific expectations in linear regression models', Behavior Research Methods.

Simple Bayesian testing of scientific expectations in linear regression models. / Mulder, Joris; Olsson Collentine, Anton.

In: Behavior Research Methods, 2019.

Research output: Contribution to journal › Article › Scientific › peer-review

TY - JOUR

T1 - Simple Bayesian testing of scientific expectations in linear regression models

AU - Mulder, Joris

AU - Olsson Collentine, Anton

PY - 2019

Y1 - 2019

N2 - Scientific theories can often be formulated using equality and order constraints on the relative effects in a linear regression model. For example, it may be expected that the effect of the first predictor is larger than the effect of the second predictor, and the second predictor is expected to be larger than the third predictor. The goal is then to test such expectations against competing scientific expectations or theories. In this paper a simple default Bayes factor test is proposed for testing multiple hypotheses with equality and order constraints on the effects of interest. The proposed testing criterion can be computed without requiring external prior information about the expected effects before observing the data. The method is implemented in R-package called `{\tt lmhyp}' which is freely downloadable and ready to use. The usability of the method and software is illustrated using empirical applications from the social and behavioral sciences.

AB - Scientific theories can often be formulated using equality and order constraints on the relative effects in a linear regression model. For example, it may be expected that the effect of the first predictor is larger than the effect of the second predictor, and the second predictor is expected to be larger than the third predictor. The goal is then to test such expectations against competing scientific expectations or theories. In this paper a simple default Bayes factor test is proposed for testing multiple hypotheses with equality and order constraints on the effects of interest. The proposed testing criterion can be computed without requiring external prior information about the expected effects before observing the data. The method is implemented in R-package called `{\tt lmhyp}' which is freely downloadable and ready to use. The usability of the method and software is illustrated using empirical applications from the social and behavioral sciences.

KW - Bayes factors

KW - regression modeling

KW - Bayesian hypothesis testing

KW - equality and order constraints

KW - Psychology

M3 - Article

JO - Behavior Research Methods

JF - Behavior Research Methods

SN - 1554-351X

ER -

Mulder J, Olsson Collentine A. Simple Bayesian testing of scientific expectations in linear regression models. Behavior Research Methods. 2019.