Predictive Analytics icon

Predictive Analytics

Run explainable forecasting operations on finance time series without arbitrary formulas.

Moving Average Forecast

AI-generated

Summary

Generate a moving average forecast for a financial time series to predict future values based on recent historical data.

Inputs

  • Currency — Optional string to specify currency metadata included in the predictive result.
  • Forecast Horizon (required) — Number of future forecast points to generate; must be >= 1.
  • Max Recommended Horizon (required) — Recommended maximum forecast horizon; horizon above this triggers warnings but is allowed.
  • Value Column (required) — Name of the column containing the numeric financial series to forecast.
  • Window Size (required) — Number of recent observations used for each moving average forecast step; must be >= 1.

Output shape

A single JSON object summarizing the forecast operation.

The output JSON includes keys: operation, valueColumn, optional currency, forecast array with future forecast points and their step number, optional warnings if horizon exceeds max recommended or if window size exceeds available data, metrics of forecasting accuracy, limitations about forecast assumptions, inputSummary describing data counts, and an auditTrail listing process steps. Errors will produce a failure output object instead.

Examples

Example 1: Forecasting the next 3 values of a financial series using a 3-point moving average.

Provide a 'valueColumn' with historical data, set 'horizon' to 3, and 'windowSize' to 3. Optionally provide 'currency' metadata and adjust 'maxHorizon' for warnings on long forecasts.

Links

Discussion