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›Finance›Johansen Cointegration Test and Vector Error Correction Model
Regression model

Johansen Cointegration Test and Vector Error Correction Model

Johansen Cointegration Test and Vector Error Correction Model (VECM) · Also known as: Johansen test, VECM, vector error correction model, multivariate cointegration, Johansen Eşbütünleşme Testi ve VECM

The Johansen procedure is a multivariate cointegration framework, introduced by Søren Johansen in 1991, that tests for long-run equilibrium relationships among several I(1) time series. It determines how many cointegrating vectors link the series and then builds a Vector Error Correction Model (VECM) to describe the short-run dynamics around that equilibrium.

ScholarGate
  1. Regression model
  2. v1
  3. 2 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.

Johansen Cointegration Test
ARDL Bounds TestARIMAVAR ModelCopula ModelsNonlinear Engle-Granger…Nonlinear Johansen Coint…Nonlinear VECMRealized VolatilityRobust ARDL bounds testRobust VECM

+3 more

When to use it

Use the Johansen test when you have several continuous time series that are each integrated of order one, I(1), confirmed by unit-root testing, and you want to know whether they share long-run equilibrium relationships. It needs a reasonably long sample (at least 50 observations, and ideally several hundred) because short series make the test unreliable. The series must not be I(2) or higher, the VAR lag length should be chosen by information criteria, and a deterministic specification should be selected from the five standard cases.

Strengths & limitations

Strengths
  • Tests for multiple cointegrating relationships at once, not just a single pair, within a full system.
  • Maximum-likelihood estimation yields the cointegrating vectors and the adjustment speeds together, and supports formal hypothesis tests on them.
  • The VECM cleanly separates long-run equilibrium from short-run dynamics, which is valuable for forecasting and structural interpretation.
Limitations
  • Unreliable on short series; with fewer than about 250 observations the ARDL bounds test is preferable.
  • Requires all series to be I(1); if any variable is I(2) or higher the test cannot be applied and differencing is needed first.
  • Results are sensitive to the chosen lag length and to the deterministic specification (trend and constant).

Frequently asked

What is cointegration?

Cointegration means that two or more individually non-stationary I(1) series move together so that a particular linear combination of them is stationary. That stable combination represents a long-run equilibrium the series keep returning to.

How is the Johansen test different from Engle-Granger?

Engle-Granger handles a single cointegrating relationship between a pair of series, while the Johansen procedure works within a full system and can detect several cointegrating vectors at once, estimating them jointly by maximum likelihood.

Trace test or maximum eigenvalue test — which should I use?

Both test the number of cointegrating vectors but in different ways and can give different answers. Report both statistics; if they disagree, examine the deterministic specification and lag choice rather than picking the convenient result.

When should I use an ARDL bounds test instead?

When the sample is short (roughly under 250 observations) the Johansen test becomes unreliable. In that case the ARDL bounds test is the preferred way to assess a long-run relationship.

Sources

  1. Johansen, S. (1991). Estimation and Hypothesis Testing of Cointegration Vectors in Gaussian Vector Autoregressive Models. Econometrica, 59(6), 1551-1580. DOI: 10.2307/2938278 ↗
  2. Johansen, S. (1995). Likelihood-Based Inference in Cointegrated Vector Autoregressive Models. Oxford University Press. ISBN: 978-0198774501

How to cite this page

ScholarGate. (2026, June 1). Johansen Cointegration Test and Vector Error Correction Model (VECM). ScholarGate. https://scholargate.app/en/finance/johansen-cointegration

Related methods

ARDL Bounds TestARIMAVAR 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.

  • ARDL Bounds TestEconometrics↔ compare
  • ARIMAEconometrics↔ compare
  • VAR ModelEconometrics↔ compare
Compare side by side →

Referenced by

ARDL Bounds TestCopula ModelsNonlinear Engle-Granger CointegrationNonlinear Johansen CointegrationNonlinear VECMRealized VolatilityRobust ARDL bounds testRobust VECMStructural break Engle-Granger cointegrationTime-varying parameter Engle-Granger cointegrationTime-varying parameter Johansen cointegration

Similar methods

Cointegration TestVECMVector Error Correction ModelPanel Johansen CointegrationStructural break Johansen cointegrationTime-varying parameter Johansen cointegrationNonlinear Johansen CointegrationEngle-Granger Cointegration Test

Related reference concepts

Mathematical and Quantitative MethodsEconometricsMultivariate Analysis of VarianceStructural Equation ModelingCanonical Correlation AnalysisMultivariate Multiple Regression

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

ScholarGate — Johansen Cointegration Test (Johansen Cointegration Test and Vector Error Correction Model (VECM)). Retrieved 2026-07-21 from https://scholargate.app/en/finance/johansen-cointegration · Dataset: https://doi.org/10.5281/zenodo.20539026
Quick facts
Originator
Søren Johansen
Year
1991
Type
Multivariate cointegration / vector error correction model
Estimator
Maximum likelihood (reduced-rank regression)
Tests
Trace and maximum eigenvalue statistics
MinSample
50
DataRequirement
Multiple I(1) (first-order integrated) time series
Related methods
ARDL Bounds TestARIMAVAR 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