Analyze regression fit with clean inputs and summaries. See diagnostics, charts, and export-ready performance details. Make smarter modeling decisions with reliable fit evidence today.
Use comma, space, or line breaks between values. The calculator compares each actual value with its matching prediction.
Here, yi is the actual value, ŷi is the predicted value, ȳ is the mean of actual values, n is the number of observations, and p is the number of predictors.
R² shows how much variability in the actual data is explained by the predictions. Adjusted R² penalizes unnecessary predictors.
| # | Actual | Predicted | Residual |
|---|---|---|---|
| 1 | 120 | 118 | 2 |
| 2 | 135 | 140 | -5 |
| 3 | 150 | 148 | 2 |
| 4 | 165 | 170 | -5 |
| 5 | 180 | 176 | 4 |
| 6 | 210 | 214 | -4 |
| 7 | 240 | 236 | 4 |
| 8 | 260 | 266 | -6 |
R² measures how much of the variation in actual values is explained by your predictions. Higher values usually indicate a better fit, but strong R² alone does not guarantee a reliable model.
Yes. A negative R² means the model performs worse than a simple baseline that always predicts the mean of the actual values. This usually signals poor fit or incorrect predictions.
Adjusted R² accounts for the number of predictors. It helps prevent misleading optimism when you add extra variables that do not truly improve model quality.
If all actual values are identical, SST becomes zero. In that case, R² is not generally informative. This calculator marks it as undefined unless the predictions are perfectly identical too.
No. Pair R² with RMSE, MAE, residual patterns, domain knowledge, and validation results. A model can show a strong R² while still making practically harmful errors.
The Plotly graph compares actual values against predicted values and adds a perfect-fit reference line. Points close to that line indicate stronger agreement between observations and predictions.
Yes. Each actual value must align with one predicted value in the same position. The calculator validates this because mismatched pairs would make the fit statistics meaningless.
You can download a CSV file containing summary metrics and row-level data. You can also generate a PDF snapshot of the calculated results for sharing or documentation.
Important Note: All the Calculators listed in this site are for educational purpose only and we do not guarentee the accuracy of results. Please do consult with other sources as well.