ExponentialSmoothingModel Class

public class ExponentialSmoothingModel : TimeSeriesModel<double>

Public Class ExponentialSmoothingModel
	Inherits TimeSeriesModel(Of Double)

public ref class ExponentialSmoothingModel : public TimeSeriesModel<double>

type ExponentialSmoothingModel = 
    class
        inherit TimeSeriesModel<float>
    end

Inheritance

Object  →  Model  →  TimeSeriesModel<Double>  →  ExponentialSmoothingModel

Use the ExponentialSmoothingModel class to remove noise or forecast data using exponential smoothing. A variety of methods is available, including single and double smoothing.

Single exponential smoothing is equivalent to computing an exponential moving average. The smoothing parameter is determined automatically, by minimizing the squared difference between the actual and the forecast values. Double exponential smoothing introduces a linear trend, and so has two parameters.

Exponential smoothing models are constructed from a Vector<T> that represents the time series data. The Method property determines the type of smoothing. It is of type ExponentialSmoothingMethod. For double exponential smoothing models, the TrendEstimator property, of type ExponentialSmoothingTrendEstimator, determines how the initial value for the trend is determined.

The Fit() method finds the parameter values that minimize the squared error of the forecast. The Forecast(Int32) method can then be used to get the one step ahead forecast, any single future forecast, or a series of forecasts as a Vector<T>.

Reference