Skip to contentScholarGate
LibraryBookshelfDeskReview StudioAssistant
Sign in
On this page
IntuitionHow it worksWhen to use itStrengths & limitationsCommon pitfallsApplicationsFrequently asked🔒 Read the full methodSourcesRelated methods
Cite this pageSpotted an issue on this page? Report or suggest a fix →
Home›Econometrics›Hausman Specification Test (FE vs RE)
Regression model

Hausman Specification Test (FE vs RE)

Hausman Specification Test (Fixed Effects vs Random Effects) · Also known as: Hausman specification test, FE vs RE test, Durbin-Wu-Hausman test, Hausman Spesifikasyon Testi (FE vs RE)

The Hausman test is a specification test, introduced by Jerry A. Hausman in 1978, that decides between the fixed-effects (FE) and random-effects (RE) estimators in panel data models. The null hypothesis is that the random-effects estimator is consistent and efficient and should be preferred; the alternative is that random effects is inconsistent and fixed effects is required because the unit-specific effects are correlated with the explanatory variables.

ScholarGate
  1. Regression model
  2. v1
  3. 1 Sources
  4. PUBLISHED
Cite this page →
Tools & resources
Download slides
Learn & explore

Read the full method

Members only

Sign in with a free account to read this section.

Sign in

Method map

The neighbourhood of related methods — select a node to explore.

Hausman Test
FMOLS EstimatorOLS RegressionPanel Cointegration TestsPanel Fixed EffectsRandom Effects ModelFourier Hausman testMundlak-ChamberlainNonlinear Hausman testRandom Effects Panel Mod…Time-varying parameter H…

When to use it

Use the Hausman test on panel (longitudinal) data with a continuous outcome once you have estimated both a fixed-effects and a random-effects model on the same specification, and you need a principled rule to choose between them. It assumes both models are correctly specified with no omitted variables, that under the null RE is consistent and efficient while FE is consistent but inefficient, and it relies on an asymptotic chi-square distribution, so a reasonable sample (at least about 50 observations) is needed. For small samples a bootstrap version of the test is recommended.

Strengths & limitations

Strengths
  • Provides a formal, widely accepted decision rule for the fixed-effects versus random-effects choice that is central to applied panel econometrics.
  • Built directly on estimators you already compute, comparing the FE and RE coefficient vectors without needing an additional model.
  • Has a clear asymptotic chi-square distribution, making the test statistic and its p-value easy to report and interpret.
Limitations
  • Relies on an asymptotic approximation, so it can be unreliable in small samples; a bootstrap Hausman test is preferable there.
  • Assumes both models are correctly specified with no omitted variables, and a rejection can reflect general misspecification rather than only correlated effects.
  • The covariance difference can be non-positive-definite in finite samples, occasionally yielding a negative or undefined statistic.

Frequently asked

What does it mean if the Hausman test rejects the null?

Rejecting the null means the random-effects estimator is inconsistent because the unit-specific effects are correlated with the regressors. In that case you should report the fixed-effects estimates, which remain consistent even when that correlation is present.

Why choose random effects if fixed effects is always consistent?

Fixed effects is consistent but discards between-unit variation and is less efficient. When the null holds, random effects is consistent and efficient, giving smaller standard errors and allowing time-invariant regressors. The Hausman test tells you whether you can safely keep that efficiency.

Is the test reliable in small samples?

Not always. The statistic follows a chi-square distribution only asymptotically, so in short panels the approximation can be poor. A bootstrap Hausman test is recommended when the sample is small, and the Mundlak (1978) approach is a more flexible alternative.

What if the test statistic comes out negative?

A negative statistic can occur in finite samples when the difference between the FE and RE covariance matrices is not positive definite. It signals that the asymptotic assumptions are strained; a bootstrap version or a robust variant of the test is the appropriate response.

Sources

  1. Hausman, J. A. (1978). Specification Tests in Econometrics. Econometrica, 46(6), 1251–1271. DOI: 10.2307/1913827 ↗

How to cite this page

ScholarGate. (2026, June 1). Hausman Specification Test (Fixed Effects vs Random Effects). ScholarGate. https://scholargate.app/en/econometrics/hausman-test

Related methods

FMOLS EstimatorOLS RegressionPanel Cointegration TestsPanel Fixed EffectsRandom Effects Model

Which method?

Set this method beside its closest kin and read them side by side — the library lays the books on the table; the choice is yours.

  • FMOLS EstimatorEconometrics↔ compare
  • OLS RegressionEconometrics↔ compare
  • Panel Cointegration TestsEconometrics↔ compare
  • Panel Fixed EffectsEconometrics↔ compare
  • Random Effects ModelEconometrics↔ compare
Compare side by side →

Referenced by

Fourier Hausman testMundlak-ChamberlainNonlinear Hausman testRandom Effects Panel ModelTime-varying parameter Hausman test

Similar methods

Panel Hausman TestRobust Hausman TestStructural Break Hausman TestRandom Effects Panel ModelFixed Effects Panel ModelBayesian Hausman TestRandom Effects ModelPanel Random Effects Model

Related reference concepts

EconometricsMultiple or Simultaneous Equation Models • Multiple VariablesStatistical Hypothesis TestingMathematical and Quantitative MethodsSingle Equation Models • Single VariablesInstrumental Variables (IV) Estimation

Spotted an issue on this page? Report or suggest a fix →

ScholarGate — Hausman Test (Hausman Specification Test (Fixed Effects vs Random Effects)). Retrieved 2026-07-21 from https://scholargate.app/en/econometrics/hausman-test · Dataset: https://doi.org/10.5281/zenodo.20539026
Quick facts
Originator
Jerry A. Hausman
Year
1978
Type
Specification test for panel data models
Estimator
Wald-type quadratic form comparing FE and RE coefficient vectors
Distribution
Asymptotic chi-square
MinSample
50
Related methods
FMOLS EstimatorOLS RegressionPanel Cointegration TestsPanel Fixed EffectsRandom Effects Model
ScholarGate

A content-first reference library for research methods — what each one is, how it works, and where it comes from.

Open data (CC-BY)

Explore

  • Library
  • Search the library…
  • Browse by field
  • Fields
  • Journey
  • Compare
  • Which method?

Reference

  • Subjects
  • Atlas
  • Glossary
  • Methodology
  • Philosophy

Your tools

  • Bookshelf
  • Desk
  • Chat

Company

  • About
  • Pricing
  • Contact
  • Suggest a method

Entries are compiled from published sources for reference. Verifying the accuracy and suitability of any information for your own use remains your responsibility.

© 2026 ScholarGate · A research-method reference library
  • Privacy
  • Cookies
  • Terms
  • Delete account