Regression error audit
Compare average signed error, absolute error and the influence of large errors on RMSE.
Your entries stay in this page’s memory and are not submitted to us. Downloads are generated locally. Reloading or leaving the page clears your entries. Do not enter patient information.
The starting values are hypothetical. Use aggregate or synthetic inputs only.
Calculation results
- Mean absolute error
- 1.6666667
- Mean signed error
- -0.33333333
- Root mean squared error
- 1.9148542
- Observations
- 3
How to interpret the output
Use frequencies of identical observed–predicted pairs on the same measurement scale. Signed error is prediction minus observation. RMSE places greater weight on large errors. Compare models on the same evaluation set.
Method reference