📊

MAPE (Mean Absolute Percentage Error) Calculator

Score a forecast against actuals — MAPE, sMAPE, MAE and bias from two paired lists.

Actual values

Forecast values (same order)

Loading calculator…

How it works

MAPE = (1/n) Σ |Aₜ − Fₜ| ÷ |Aₜ| × 100

MAPE expresses forecast error as a percentage of the actual, which makes it easy to explain but unstable when actuals are near zero and asymmetric (over-forecasts are penalised less). sMAPE divides by the average of actual and forecast to soften both problems; MAE keeps the original units.

Worked example

Actuals 120…170 vs forecasts 110…165: absolute percentage errors of 8.3%, 3.7%, 3.3%, 6.3%, 3.2%, 2.9% → MAPE = 4.6%.

Frequently asked questions

What is a good MAPE?

Under 10% is excellent for most business forecasts; 10–20% good; over 50% means the model is not useful (Lewis, 1982).

Why is MAPE undefined?

Any actual equal to zero makes its term divide by zero; those periods are skipped in MAPE but included in sMAPE and MAE.

Related calculators

RMSE (Root Mean Square Error) Calculator

Compute RMSE, MSE, MAE and R² between predicted and observed values from two paired lists.

Percent Error Calculator

Find the percent error between a measured value and the accepted true value with the working shown, plus absolute, signed and relative error. Average several lab trials, check a tolerance, or work back from a percent error.

Moving Average Calculator

Smooth a series with a simple moving average — pick a window size and see the averaged values that reveal the underlying trend.

For education — check assumptions (sample size, distribution) before using statistics in real decisions.