Definition

Problem in regression models where predictor variables are correlated with the error term , i.e.,

Common causes

  • Omitted variable bias: There is an important factor that is not included in the model.
  • Simultaneity: Bidirectional relationship (e.g., cost & demand influences each other).
  • Measurement error: is measured with high error.

Consequences

OLS estimator becomes biased and inconsistent. Interpretation of coefficient could be incorrect.