Package

weka.classifiers.functions

Synopsis

Implements a least median sqaured linear regression utilising the existing weka LinearRegression class to form predictions.
Least squared regression functions are generated from random subsamples of the data. The least squared regression with the lowest meadian squared error is chosen as the final model.

The basis of the algorithm is

Peter J. Rousseeuw, Annick M. Leroy (1987). Robust regression and outlier detection.

Available in Weka 3.6.x - 3.7.1. Available via the package management system for Weka >= 3.7.2 (leastMedSquared).

Options

The table below describes the options available for LeastMedSq.

Option	Description
debug	If set to true, classifier may output additional info to the console.
randomSeed	Set the seed for selecting random subsamples of the training data.
sampleSize	Set the size of the random samples used to generate the least sqaured regression functions.

Capabilities

The table below describes the capabilites of LeastMedSq.

Capability	Supported
Class	Date class, Numeric class, Missing class values
Attributes	Binary attributes, Nominal attributes, Numeric attributes, Unary attributes, Missing values, Date attributes, Empty nominal attributes
Min # of instances	1