In this section, we describe some of the sample applications included with the OpenForecast source code.
This example application demonstrates a bare bones, console-based application that uses the steps described in the section called Writing an Application using OpenForecast to forecast some future values for a simple series. This small application may be a good starting point when writing your first OpenForecast application.
This sample application is a more involved example that demonstrates the use of OpenForecast in conjunction with the JFreeChart charting library to produce a chart of the observations along with forecast values for a variety of different forecasting models. This application also illustrates how to select and use specific forecasting models.
The ExponentialSmoothingChartDemo is similar to the ForecastingChartDemo in that it makes use of JFreeChart to illustrate a data series alongside forecast values. However, the models used by this application are specifically single, double and triple exponential smoothing applied to a data set that illustrates both trend and seasonality.
The example shows how single exponential smoothing is really not a good fit for this kind of data, how double exponential smoothing is better by adding the trend, and how triple exponential exponential smoothing is most appropriate by adding the seasonality as well as the trend.
Note that, as of the time of this writing, this application only illustrates single exponential smoothing. This is because double and triple exponential smoothing models have not yet been implemented in OpenForecast. Once support for these models is added, this application should be updated to illustrate these models, as described in this section.