Automatic Parameter Search

Click the Automatic estimation button on the Seasonal and Non-Seasonal Exponential Smoothing dialog box - Automatic Search tab (after specifying the criteria for the automatic search) to perform the procedure. After the estimation procedure has been performed, the analysis results are displayed. If the parameter estimation procedure fails to converge, the Automatic Parameter Search dialog is displayed and you have the option to request additional iterations.

The parameter estimation procedure is iterative in nature, that is, parameter estimates are refined in successive iterations. A quasi-Newton algorithm is used to minimize the selected Lack of fit indicator. The respective value of the indicator is listed in the Loss column in the Automatic Parameter Search dialog; the parameter values are listed to the right.   

Note that, by default (when the respective option is checked on the Automatic Search tab) the parameter search is unconstrained, that is, parameter values outside the 0/1 boundaries may be estimated.  After the parameter search converges, invalid parameters will automatically be set to the respective minimum or maximum (e.g., a = -0.2 is set to a = 0; a = 1.2 is set to a = 1); and the a full set of results will be displayed for the best parameters.