Kostenloser Adjusted R-Squared Rechner
Berechnen Sie adjusted R² to compare regression models mit different numbers of predictors. Avoid overfitting.
Adjusted R²
0.8402
Adjusted R² vs R²
Formel
## Adjusted R-Squared Adjusted R² modifies R² to penalize for additional predictors that do not improve the model. Unlike R², it can decrease when unhelpful variables are added. ### Formula **Adj R² = 1 - (1 - R²) × (n - 1) / (n - p - 1)** where n is the sample size and p is the number of predictors. Adjusted R² is always less than or equal to R². The difference grows larger with more predictors relative to sample size.
Lösungsbeispiel
R² = 0.85, n = 50, p = 3 predictors.
- 01Adj R² = 1 - (1 - 0.85) × 49/46
- 02Adj R² = 1 - 0.15 × 1.0652 = 1 - 0.1598 = 0.8402
- 03Penalty = 0.85 - 0.84 = 0.01
Häufig Gestellte Fragen
When should I use adjusted R² instead of R²?
Always use adjusted R² when comparing models with different numbers of predictors. Regular R² is misleading because it can only increase with more variables, even useless ones.
Can adjusted R² be negative?
Yes. If the model fits worse than just using the mean (e.g., too many useless predictors with small n), adjusted R² can go below zero. This indicates the model is not useful.
How does AIC/BIC compare to adjusted R²?
AIC and BIC are alternative model selection criteria based on likelihood with different penalty terms. They can be used for non-linear models too. Adjusted R² is simpler but limited to comparing linear models.
Lernen