A simple moving average forecast uses the most recent observations to estimate the next period. Enter demand in chronological order, choose a window, and this tool separates the next-period forecast from historical rolling fitted values and forecast errors.
Moving Average Forecast Calculator
Transparent baseline forecasting for ordered demand data
How a simple moving average forecast works
For a window of k periods, the forecast for period t is the arithmetic mean of the previous k actual observations. With demand of 100, 120, 140 and 160 and a 3-period window, the next-period forecast is (120 + 140 + 160) ÷ 3 = 140.
The tool also backtests the method wherever enough history exists. It reports MAE and RMSE from those one-step-ahead historical forecast errors. These diagnostics describe performance on the supplied history; they do not guarantee future accuracy.
When should you use a moving average?
A simple moving average is useful as a transparent baseline when observations are ordered at equal intervals and the underlying level is reasonably stable. It smooths short-term noise, but it tends to lag sustained trends and does not explicitly model recurring seasonality, promotions, holidays or other causal drivers.
If demand has strong trend or seasonal structure, compare this baseline with methods designed for those patterns rather than assuming a larger moving-average window will solve the problem.
Choosing the window
A shorter window responds faster to recent changes but is usually noisier. A longer window smooths more aggressively but reacts more slowly. Select the window based on the business process and evaluate forecast error on history. Avoid choosing a window only because it produces the most visually pleasing line.
Related tools
Use the XYZ Analysis Calculator to examine relative historical demand variability. For replenishment decisions, continue to the Safety Stock & Reorder Point Calculator or EOQ Calculator.