R Squared

R-squared (R²) is a measure of how much variability in the dependent variable is explained by the independent variables in a regression model. It is a coefficient of determination that ranges from 0 to 1, with values closer to 1 indicating a better fit of the model to the data.

Formula:

R² = 1 - (SSres / SStotal)

where:

• R² is the coefficient of determination
• SSres is the sum of squares of residuals
• SStotal is the total sum of squares

Interpretation:

• R² = 0: The model does not explain any variability in the dependent variable.
• R² = 1: The model explains all variability in the dependent variable.
• 0 < R² < 1: The model explains a certain percentage of variability in the dependent variable.

Example:

R² = 0.85

This means that the model explains 85% of the variability in the dependent variable.

Uses:

• To assess the overall fit of a regression model.
• To compare different models and select the best-fitting model.
• To evaluate the predictive power of a model.

Limitations:

• Overfitting: If the model is too complex, it can lead to an inflated R-squared value.
• Sensitive to outliers: Outliers can significantly impact R-squared values.
• Not a perfect measure: R-squared does not necessarily reflect the accuracy of a model’s predictions.

Additional Notes:

• R-squared is a commonly used metric in regression analysis.
• It is a popular measure of model fit, but it should not be the only factor considered when selecting a model.
• Other factors, such as the model’s complexity and the presence of outliers, should also be taken into account.