Function forecasting::ExponentialSmoothingTune(dataValues, noObservations, alpha, alphaLow, alphaUpp)


The forecasting::ExponentialSmoothingTune() procedure is a time series forecasting helper procedure of forecasting::ExponentialSmoothing() by computing the \(\alpha\) for which the mean squared error is minimized.

Function Prototype

! Provides the alpha for which the mean squared error is minimized.
        dataValues,      ! Input, parameter indexed over time set
        noObservations,  ! Scalar input, length history
        alpha,           ! Scalar output, weight of observation
                        ! that minimizes mean squared error
        alphaLow,        ! Optional input, default 0.01
        alphaUpp)        ! Optional input, default 0.99



A one dimensional parameter containing the observations for the first \(T\) elements of the time set.


Specifies the number of elements that belong to the history of the time set. This parameter corresponds to \(T\) in the notation presented in Time Series Forecasting Notation.


Upon return it provides the weighting factor \(\alpha\) for which the mean squared error is minimized when using forecasting::ExponentialSmoothing() on the same dataValues.


Lowerbound on \(\alpha\), default 0.01.


Upperbound on \(\alpha\), default 0.99.


In order to use this function, the Forecasting system library needs to be added to the application.

Please note that this function performs an optimization step; a nonlinear programming solver should be available and, in an AIMMS PRO environment, it should be run server side.


To further understand about this procedure and library, please use the Demand Forecasting example.