The Analytics.h++ library consists of three main components:
the regression classes
the parameter calculation classes
the model selection classes
The parameter calculation classes calculate regression parameters from input regression data. The regression classes contain regression data and a parameter calculation object. These classes couple regression data and parameter calculations; the currently calculated parameters and the regression data contained by a regression object are always in sync. Finally, the model selection classes take a regression object from which they extract the regression data and calculation object. They search subsets of the predictors and select an optimal subset by applying a user-defined model evaluation function.
©Copyright 1999, Rogue Wave Software, Inc.
Contact Rogue Wave about documentation or support issues.