com.numericalmethod.suanshu.stats.regression.linear.glm

Class GLMProblem

java.lang.Object

com.numericalmethod.suanshu.stats.regression.linear.LMProblem

com.numericalmethod.suanshu.stats.regression.linear.glm.GLMProblem

All Implemented Interfaces:

java.io.Serializable

Direct Known Subclasses:

QuasiGlmProblem

public class GLMProblem
extends LMProblem

This class represents a Generalized Linear regression problem.

See Also:

Serialized Form

Field Summary

Fields

Modifier and Type	Field and Description
Family	family the exponential family distribution for the mean

Fields inherited from class com.numericalmethod.suanshu.stats.regression.linear.LMProblem

A, addIntercept, wA, weights, wy, y

Constructor Summary

Constructors

Constructor and Description
GLMProblem(LMProblem problem, Family family) Construct a GLM problem from a linear regression problem.
GLMProblem(Vector y, Matrix X, boolean addIntercept, Family family) Construct a GLM problem.

Method Summary

Methods inherited from class com.numericalmethod.suanshu.stats.regression.linear.LMProblem

checkInputs, invOfwAtwA, nExogenousFactors, nFactors, nObs, X

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Field Detail

family

public final Family family

the exponential family distribution for the mean

Constructor Detail

GLMProblem

public GLMProblem(Vector y,
                  Matrix X,
                  boolean addIntercept,
                  Family family)

Construct a GLM problem.

Parameters:

y - the dependent variables

X - the factors

addIntercept - true if to add an intercept term to the linear regression

family - the exponential family distribution for the mean

GLMProblem

public GLMProblem(LMProblem problem,
                  Family family)

Construct a GLM problem from a linear regression problem.

Parameters:

problem - a linear regression problem

family - the exponential family distribution for the mean