Analytics Magazine Analytics Magazine, July/August 2014 | Page 78

FO RE CAST ING Figure 6: IBM SPSS’ fitted models for three specified growth curves. IBM SPSS does have a curve fitting feature, so I utilized it and specified three possible models to be examined – the linear, growth and logistic curves. Figures 6 and 7 give the resulting output and plots for these choices. NCSS has, in addition to the stanFigure 7: IBM SPSS’ plot of the data and growth curves. dard forecasting procedures (Box-Jenkins and exponen- with four parameters; there is a Logistic(3) tial smoothing models), an extensive list model available as well], and Figure 8 of more than 20 nonlinear curve mod- shows the NCSS’ output. els under its menu label “Growth and Minitab is a hybrid of a semi-autoOther Models.” The user chooses a matic and manual forecasting product. If model, and NCSS finds the appropriate you specify that a Box-Jenkins model be parameters for the particular data set. used, the software finds the appropriate I chose, for the “Non-Stop” data, the parameters for the model. However, if you “Logistic(4)” model [i.e., a logistic curve choose Winters’ method, Minitab requires 78 | A N A LY T I C S - M A G A Z I N E . O R G W W W. I N F O R M S . O R G