The Engle-Granger Method

The Engle-Granger method is a powerful econometric technique that enables economists and researchers to analyze and model the long-term equilibrium relationships between variables, known as cointegration.



2/5/20243 min read

chart trading
chart trading

The Engle-Granger method is an econometric technique that has gained significant importance in the analysis of financial and macroeconomic time series. Developed by Robert Engle and Clive Granger in the 1980s, this method allows economists and researchers to analyze and model the long-term equilibrium relationship between two variables, also known as cointegration.

What is Cointegration?

Cointegration is a statistical property that exists when two or more time series are tied together in the long run, despite exhibiting short-term fluctuations. In other words, it refers to the presence of a stable relationship between variables that may have a non-stationary behavior individually.

For example, consider the relationship between the prices of two stocks. While the prices of individual stocks may fluctuate in the short term, they may exhibit a long-term equilibrium relationship due to factors such as market trends or industry-specific factors. Cointegration analysis helps identify and model such relationships.

The Engle-Granger Method: An Overview

The Engle-Granger method provides a framework for testing and estimating cointegration between two variables. The process involves three main steps:

  1. Step 1: Test for Stationarity

    The first step in the Engle-Granger method is to test whether the individual time series are stationary or not. Stationarity refers to the property of a time series where its statistical properties, such as mean and variance, remain constant over time. Non-stationary time series may exhibit trends, cycles, or other forms of systematic patterns.

    There are various statistical tests available to determine the stationarity of a time series, such as the Augmented Dickey-Fuller (ADF) test or the Phillips-Perron (PP) test. If the individual series are found to be non-stationary, the analysis proceeds to the next step.

  2. Step 2: Estimate the Cointegration Relationship

    In this step, the Engle-Granger method estimates the long-term equilibrium relationship between the variables by regressing one variable on the other, typically using ordinary least squares (OLS) regression. The residual series from this regression represents the difference between the actual and predicted values of the dependent variable.

    If the residual series is found to be stationary, it indicates the presence of cointegration between the variables. In other words, the long-term relationship between the variables can be captured by a stationary linear combination of the two series.

  3. Step 3: Test the Residuals for Stationarity

    The final step involves testing the stationarity of the residual series obtained in the previous step. If the residuals are found to be stationary, it confirms the existence of a cointegrating relationship between the variables.

    Various tests can be used to assess the stationarity of the residuals, such as the ADF test or the PP test. If the residuals are stationary, the cointegration analysis is considered successful, and the estimated relationship can be used for further analysis and modeling.

Applications of the Engle-Granger Method

The Engle-Granger method has found widespread applications in various fields, particularly in finance and macroeconomics. Some of its key applications include:

1. Financial Market Analysis

The Engle-Granger method is extensively used in the analysis of financial time series, such as stock prices, exchange rates, and interest rates. By identifying cointegrating relationships between these variables, economists and analysts can gain insights into long-term equilibrium relationships and potential trading strategies.

For example, in the case of pairs trading, where two stocks are identified as cointegrated, traders can take advantage of temporary deviations from the estimated equilibrium relationship to generate profitable trading opportunities.

2. Macroeconomic Analysis

In macroeconomics, the Engle-Granger method is employed to analyze the long-term relationships between macroeconomic variables, such as GDP, inflation, and unemployment. By understanding the cointegrating relationships, policymakers can make informed decisions regarding economic stability, monetary policy, and fiscal measures.

For instance, if a cointegrating relationship is found between inflation and money supply, policymakers can use this information to guide their decisions on interest rates and money supply adjustments to maintain price stability.

3. Forecasting and Model Building

The Engle-Granger method provides a foundation for building econometric models that capture the long-term relationships between variables. These models can then be used for forecasting and scenario analysis.

By incorporating cointegration analysis into forecasting models, economists can improve the accuracy of their predictions, especially for variables that exhibit long-term relationships. This is particularly valuable in industries such as energy, where the prices of commodities like oil and gas are influenced by long-term market dynamics.


The Engle-Granger method is a powerful econometric technique that enables economists and researchers to analyze and model the long-term equilibrium relationships between variables, known as cointegration. By following a systematic approach of testing for stationarity, estimating the cointegration relationship, and testing the residuals for stationarity, this method provides valuable insights into the dynamics of time series data.

With its applications in financial market analysis, macroeconomic analysis, and forecasting, the Engle-Granger method has become an indispensable tool for understanding and modeling complex economic relationships. By leveraging this technique, economists can make more informed decisions and predictions, contributing to the advancement of economic theory and practice.

You might be interested in