Forecasts future values for a time series of dated observations, using openforecast's double exponential smoothing model. Normally obtained via RegressionBuilder.buildForecaster, which constructs and trains the model in one step. Call predictions() or timeSeries() afterwards to retrieve the forecast, one point per period from the last observation up to the configured end date.

Properties

Methods

getName()

Returns: String

Name of this model, inherited from the RegressionBuilder it was built from. Used only to identify the model in log messages and has no effect on forecasting.

train()

Returns: void

Trains the underlying double exponential smoothing forecaster against the observed data points. Called automatically by RegressionBuilder.buildForecaster; safe to call again to retrain from the same observations.

predictions()

Returns: List<DateDataPoint>

Computes the forecast, one point per period from the last observation up to predictionEndDate, and returns the predicted points. Computed once, on first call, and cached for subsequent calls. Each returned point is a DateDataPoint with isPredicted() set to true.

getPredictedValue()

Returns: double

Value forecast for the last predicted period, i.e. the point at predictionEndDate. Triggers prediction if it has not already run. Throws a NullPointerException if predictionEndDate is not after the last observation's date, since no periods are forecast in that case.

timeSeries()

Returns: List<DateDataPoint>

Actual and predicted values combined into a single chronological time series: this model's original observations followed by the forecast produced by predictions().