Skip to content
Econometrics and Financial Time Series / Nonstationarity, Unit Roots, and Cointegration

Error-Correction Models: Short-Run Dynamics and Long-Run Equilibrium

A pure return regression is safe for I(1) data but silent about whether a spot--futures basis or relative price has drifted too far from equilibrium. An error-correction model (ECM) retains short-run changes and feeds yesterday’s stationary disequilibrium into today’s response.

Learning objectives. By the end of this note you should be able to:
  • write and interpret a bivariate ECM;
  • distinguish short-run coefficients from an adjustment speed;
  • state why the equilibrium error must be lagged and stationary;
  • connect a bivariate ECM to ARDL and VECM representations.

In this lesson

  1. The equilibrium error restores the missing information
  2. From ARDL to ECM

Before you start

The full lesson, with worked examples and exercise solutions, is for members.

Read the full lesson

New to SelfMadeQuant.com? See membership options or browse the Econometrics and Financial Time Series syllabus.