Spot forecast bias before it distorts inventory decisions. Review errors, totals, and accuracy metrics instantly. Make confident forecasting decisions using transparent calculations and summaries.
Paste values separated by commas, spaces, or new lines.
| Period | Actual | Forecast | Error |
|---|---|---|---|
| Jan | 120 | 118 | 2 |
| Feb | 128 | 130 | -2 |
| Mar | 133 | 129 | 4 |
| Apr | 141 | 145 | -4 |
| May | 150 | 147 | 3 |
| Jun | 158 | 154 | 4 |
Period Error: Errort = Actualt − Forecastt
Cumulative Forecast Error: CFE = Σ(Actualt − Forecastt)
Mean Error: ME = CFE ÷ n
Mean Absolute Deviation: MAD = Σ|Errort| ÷ n
Tracking Signal: TS = CFE ÷ MAD
Root Mean Squared Error: RMSE = √(ΣErrort2 ÷ n)
MAPE: Average of |Errort| ÷ Actualt × 100 for periods where actual values are not zero.
A positive CFE usually indicates underforecasting. A negative CFE usually indicates overforecasting.
It measures the running total of forecast errors across periods. The metric shows whether forecasts systematically drift above or below actual results over time.
This calculator uses actual minus forecast. Positive values mean actual demand exceeded the forecast. Negative values mean the forecast was higher than actual demand.
Tracking signal compares cumulative forecast error to mean absolute deviation. It helps reveal persistent bias that might be hidden when you only inspect isolated period errors.
A positive cumulative forecast error usually means underforecasting. Actual values are collectively higher than forecast values, which can signal stockout risk or unmet demand.
A negative cumulative forecast error usually means overforecasting. Forecasts are collectively higher than actual values, which can suggest excess inventory or inflated demand expectations.
Yes. The calculator accepts integers and decimals, so it works for units, currency, volumes, energy demand, and many other forecasting datasets.
MAPE needs non-zero actual values. When actual values are zero, percentage error cannot be computed for those periods, so the calculator excludes them.
Investigate when the tracking signal breaches your chosen control limit, when cumulative error keeps drifting, or when operational decisions are affected by repeated bias.
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.