Multicollinearity occurs when

Multicollinearity occurs when

• A. Two or more independent variables in a regression model are highly correlated with each other.
• B. The dependent variable is highly correlated with the error term.
• C. The residuals are heteroscedastic.
• D. The model has too many observations.

Answer: (A)

Multicollinearity occurs when two or more independent variables in a regression model are highly correlated with each other. This overlap makes it hard to separate each predictor’s unique contribution to the outcome, leading to inflated standard errors and unstable coefficient estimates across samples. The model’s overall fit may look fine, but the individual predictors become difficult to interpret because their effects are entangled. This issue is not about the dependent variable being correlated with the error term (that’s endogeneity) or about the residuals having non-constant variance (heteroscedasticity). Having more observations doesn’t cause multicollinearity; in fact, more data can help identify the separate effects when predictors aren’t too similar. In practice, you’d diagnose with metrics like the variance inflation factor and address it by removing or combining collinear predictors or using methods such as ridge regression to stabilize estimates.

Multicollinearity occurs when two or more independent variables in a regression model are highly correlated with each other. This overlap makes it hard to separate each predictor’s unique contribution to the outcome, leading to inflated standard errors and unstable coefficient estimates across samples. The model’s overall fit may look fine, but the individual predictors become difficult to interpret because their effects are entangled.

This issue is not about the dependent variable being correlated with the error term (that’s endogeneity) or about the residuals having non-constant variance (heteroscedasticity). Having more observations doesn’t cause multicollinearity; in fact, more data can help identify the separate effects when predictors aren’t too similar. In practice, you’d diagnose with metrics like the variance inflation factor and address it by removing or combining collinear predictors or using methods such as ridge regression to stabilize estimates.