Class RWLinearRegression solves the multiple linear regression problem described in Multiple Linear Regression and class RWLogisticRegression solves the logistic regression problem described in Logistic Regression.

The public interfaces for these classes have much in common since they both derive from the base class, RWRegression. Both of these classes accept regression data in the form of a matrix that contains predictor data, and a vector that contains observation data. After accepting and processing the data, the classes can provide parameter estimates and their associated statistics, dispersion matrices, residuals, and variances.

Instances of RWLinearRegression may be used as input to other analysis classes. Class RWLinearRegressionANOVA takes an instance of RWLinearRegression and computes a variety of analysis of variance quantities, including overall F statistic, coefficient of determination and adjusted coefficient of determination, mean square error, and degrees of freedom. Given an instance of RWLinearRegression, class RWLinearRegressionFTest may be used to test hypotheses about the model parameters.

Similarly, RWLogisticRegression can be used as an input for analyses. For example, given an instance of RWLogisticRegression, RWLogisticFitAnalysis computes such quantities as deviance, G statistic, log likelihood, the Pearson statistic, and the Hosmer-Lemeshow statistic.