Computing & Software
Prediction MAE, RMSE, Bias & R-Squared Calculator
Evaluate existing numeric predictions against observations without fitting a new model.
How the calculation works
Mean absolute error = prediction_mae(actual,predicted) input units. Root mean squared error = prediction_rmse(actual,predicted) input units. Mean signed prediction bias = prediction_bias(actual,predicted) input units. R-squared against observed-mean baseline = prediction_r2(actual,predicted) R².
Worked example
Example inputs — Actual values: 1, 2, 3; Predicted values in matching order: 1, 2, 4. Expected results — Mean absolute error: 0.3333333333 input units; Root mean squared error: 0.5773502692 input units; Mean signed prediction bias: 0.3333333333 input units; R-squared against observed-mean baseline: 0.5 R².
Assumptions and limits
Analytical software/hardware scenario, not a benchmark or a live provider quote. Use the actual architecture, workload and billing definitions; overheads not named in the model are excluded. Equal-length paired lists with nonconstant actual values for R². Negative R² is possible for poor predictions. This differs from fitting a regression line to the same observations.
Reference and further reading
Read the underlying subject reference. Calculith is an independent implementation; the reference does not endorse or certify this tool.
Related calculators
- Password Entropy & Brute-Force Time Estimate
- CPU Instruction Count & CPI Runtime Calculator
- Amdahl & Gustafson Parallel Speedup Calculator
- Roofline Arithmetic Intensity Performance Calculator
- Memory Bus Theoretical Bandwidth Calculator
Example values are editable. Results are estimates where assumptions apply. Calculations run in your browser. Updated 2026-11-16.