Forecast Error Input
Enter actual and predicted values row by row. Zero actual values are excluded because percentage error becomes undefined.
Example Data Table
This sample shows six periods with actual values, predicted values, and the resulting percentage errors used in the MAPE summary.
| Period | Actual | Predicted | Absolute Error | APE % |
|---|---|---|---|---|
| Jan | 120 | 114 | 6 | 5.00% |
| Feb | 135 | 140 | 5 | 3.70% |
| Mar | 128 | 134 | 6 | 4.69% |
| Apr | 150 | 142 | 8 | 5.33% |
| May | 160 | 171 | 11 | 6.88% |
| Jun | 145 | 138 | 7 | 4.83% |
For these six rows, the mean of the APE values is approximately 5.07%, which indicates a strong forecasting fit.
Formula Used
MAPE stands for Mean Absolute Percentage Error. It measures the average percentage difference between actual values and predicted values.
APE = |Actual - Predicted| / |Actual| x 100
MAPE = (Sum of all valid APE values) / n
Where:
- Actual is the observed value.
- Predicted is the forecasted value.
- APE is the absolute percentage error for one row.
- n is the number of rows with nonzero actual values.
If an actual value equals zero, the percentage error is undefined. This calculator flags such rows and excludes them from the final MAPE calculation.
How to Use This Calculator
- Enter a dataset name and optional unit label.
- Set the decimal precision and your preferred interpretation thresholds.
- Fill each row with a label, actual value, and predicted value.
- Use Load Example if you want a quick sample.
- Press Calculate MAPE to display the result section above the form.
- Review row level errors, excluded rows, and the accuracy interpretation.
- Download the detailed rows as CSV or export the visible page as PDF.
Why Analysts Use MAPE
MAPE is popular because it expresses error in percentage terms, making results easy to explain across teams. It works best when actual values are positive, stable, and not close to zero.
FAQs
1. What does MAPE measure?
MAPE measures the average absolute percentage difference between actual and predicted values. It shows how far forecasts deviate from observations in percentage terms.
2. Why are zero actual values excluded?
When the actual value is zero, percentage error becomes undefined because the formula divides by the actual value. Excluding those rows prevents misleading results.
3. Is a lower MAPE always better?
Yes. Lower MAPE means forecasts are closer to actual observations on average. A smaller percentage error generally indicates better predictive accuracy.
4. What is considered a good MAPE?
Interpretation depends on the industry and data pattern. Many teams view values under 10% as strong, 10% to 20% as fair, and above 20% as weaker.
5. Can MAPE compare two forecasting models?
Yes. If both models use the same dataset, MAPE helps compare average percentage error. The model with the lower MAPE usually fits better.
6. Does MAPE work with negative values?
It can be computed, but interpretation may become less intuitive. MAPE is most reliable when actual values are positive and not close to zero.
7. What is the difference between MAPE and MAE?
MAE reports average absolute error in original units, while MAPE reports average absolute error as a percentage. MAPE is easier for relative comparisons.
8. Why export CSV or PDF results?
Exports make it easier to share row level calculations, archive model checks, and include summary metrics in audit trails, presentations, or performance reports.