OpenForecast free download, OpenForecast download on software download

Free and open source forecasting models written in Java OpenForecast is a package of general purpose, forecasting models written in Java that can be applied to any data series. One of the design goals is to make it easy for a developer to use in an application even if they do not understand, or care to understand, the differences between the different forecasting models available.NOTE: OpenForecast is licensed and distributed under the terms of the GNU Library or Lesser General Public License (LGPL). Requirements: · J2SDK 1.3 or later · Ant 1.5 or later What's New in This Release: · Modified source code distribution (defined by the "dist" target in build.xml) to include required library files. · Updated Forecaster class to consider Double and Triple Exponential Smoothing. · Re-organized test and examples source files, as well as build directories. Updated build.xml to support new structure. · Updated build.xml to separate out the build of the main library classes from the UI classes (still under development, and not part of the 0.4 release), the test cases and examples. · Updated build.xml to compile TimeSerieshelper classes (require JFreeChart.jar). Actually modified all targets to use a common CLASSPATH. · Added TimeSeriesBuilder and TimeSeriesOutputter (and a simple test cases for each). · Added a triple exponential smoothing model. · Updated examples to use JFreeChart 0.9.16 and JCommon 0.9.1. · Added the static SimpleExponentialSmoothing.getBestFitModel methods to select "best fit" value of alpha for a given data set. · Modified SimpleExponentialSmoothing.forecast to handle future time periods by extrapolation. · Added the static DoubleExponentialSmoothing.getBestFitModel methods to select "best fit" values of alpha and gamma for a given data set. · Modified DoubleExponentialSmoothing.forecast to handle future time periods by extrapolation. · Updated ExponentialSmoothingChartDemo to include forecast generated from a DoubleExponentialSmoothingModel model. · Added DoubleExponentialSmoothingModel extends AbstractTimeBasedModel. · Refactored SimpleExponentialSmoothingModel and WeightedMovingAverageModel to subclass new abstract class, AbstractTimeBasedModel. · Added Util.calculateSeasonalIndices() helper method to help with calculation of seasonal indices. · Added testContains, testHashCode, testRemove and testRetainAll to DataSetTest, and re-organized tests to make use of a common setUp and tearDown. · Modified DataSet to fully implement the java.util.Collection interface. So, DataSet now "is a" Collection. · Create new test class, DataSetTest and added testDataSetEqualsMethod. Moved the DataSet tests from BasicTest into this new class. · Implemented DataSet.equals(DataSet). · Implemented overridden equals method for DataSet. · Added a new package, org.sourceforge.openforecast.output, that will contain helper "outputter" classes. Defined Outputter interface, and implemented an initial DelimitedTextOutputter to output values from a DataSet to a delimited text file such as CSV file. · Added a new package, org.sourceforge.openforecast.input, that will contain helper "builder" classes. Defined Builder interface, and implemented an initial CSVBuilder to read CSV files and create a DataSet from the values. Also defined the outline to a ResultSetBuilder class. · Added getIntercept and getCoefficient methods to MultipleLinearRegressionModel class. · Added an SimpleExponentialSmoothingModel as subclass of WeightedMovingAverageModel. · Modified Observation.getIndependentValue to throw an IllegalArgumentException if an attempt was made to reference an independent variable unknown to the Observation. · Added sample of using OpenForecast with JFreeChart. · Modified build.xml to build example programs in build/classes. · Modified MovingAverageModel to subclass WeightedMovingAverageModel. · Implemented WeightedMovingAverageModel. · Modified to include support for J2SDK 1.3.

OpenForecast Related Software