org.statcato.statistics.inferential

Class MultipleRegression2

java.lang.Object

org.statcato.statistics.inferential.MultipleRegression2

public class MultipleRegression2
extends java.lang.Object

Multiple regression for non-linear models. Variations (explained, unexplained, total) are calculated using matrix operations.

Since:

1.0

Constructor Summary

Constructors

Constructor and Description
MultipleRegression2(java.util.Vector<java.util.Vector<java.lang.Double>> IndependentVars, java.util.Vector<java.lang.Double> DependentVar, boolean hasConstant) Constructor.

Method Summary

Modifier and Type	Method and Description
double	AdjustedCoefficientOfDetermination() Returns the adjusted coefficient of determination.
double	CoefficientOfDetermination() Returns the coefficient of determination r^2, the amount of the variation in y that is explained by the regression line.
double	ExplainedVariation() Returns the explained variation (SSR, the sum of squared differences between the predicted y value and the average y value).
int	NumIndepVar() Returns the number of independent variables.
double	PValue() Returns the p-Value of test statistics.
Matrix	RegressionEqCoefficients() Returns the coefficients of the regression equation y = b_0 + b_1 * x_1 + ...
int	SampleSize() Returns the sample size.
double	StandardError() Returns the standard error of estimate, sqrt(unexplained variation / (n-2)).
double	TestStatistics() Returns the test statistics F.
java.lang.String	toString()
double	TotalVariation() Returns the total variation (SST, the sum of squared differences between the y values and the average y value).
double	UnexplainedVariation() Returns the unexplained variation (the sum of squared differences between the predicted y value and the y value).
Matrix	XVar(int i) Returns the i th value of all the independent variables as a 1 by k matrix.
double	YPredicted(Matrix var) Returns the predicted y value given a vector of values of the independent variables using the regression equation.

Methods inherited from class java.lang.Object

equals, getClass, hashCode, notify, notifyAll, wait, wait, wait

Constructor Detail

MultipleRegression2

public MultipleRegression2(java.util.Vector<java.util.Vector<java.lang.Double>> IndependentVars,
                           java.util.Vector<java.lang.Double> DependentVar,
                           boolean hasConstant)

Constructor.

Parameters:

IndependentVars - a vector of vectors of double, where each vector is an independent variable with the same number of values

DependentVar - a vector of double, which has the same number of values as the independent variables

Method Detail

RegressionEqCoefficients

public Matrix RegressionEqCoefficients()

Returns the coefficients of the regression equation y = b_0 + b_1 * x_1 + ... + b_k * x_k as a matrix of dimension k+1 by 1: [b_0 b_1 ... b_k]'.

Returns:

a k+1 by 1 matrix containing the coefficients of the regression equation

YPredicted

public double YPredicted(Matrix var)

Returns the predicted y value given a vector of values of the independent variables using the regression equation.

Parameters:

var - a vector of double that has the same number of values as the number of independent variables.

Returns:

predicted y value

TotalVariation

public double TotalVariation()

Returns the total variation (SST, the sum of squared differences between the y values and the average y value).

Returns:

total variation

XVar

public Matrix XVar(int i)

Returns the i th value of all the independent variables as a 1 by k matrix.

Parameters:

i - index

Returns:

matrix

ExplainedVariation

public double ExplainedVariation()

Returns the explained variation (SSR, the sum of squared differences between the predicted y value and the average y value).

Returns:

explained variation

UnexplainedVariation

public double UnexplainedVariation()

Returns the unexplained variation (the sum of squared differences between the predicted y value and the y value).

Returns:

unexplained variation

CoefficientOfDetermination

public double CoefficientOfDetermination()

Returns the coefficient of determination r^2, the amount of the variation in y that is explained by the regression line.

Returns:

r^2

AdjustedCoefficientOfDetermination

public double AdjustedCoefficientOfDetermination()

Returns the adjusted coefficient of determination.

Returns:

r^2

StandardError

public double StandardError()

Returns the standard error of estimate, sqrt(unexplained variation / (n-2)).

Returns:

s

TestStatistics

public double TestStatistics()

Returns the test statistics F. F = (explained variation) / (unexplained variation) * (n - k - 1) / k

Returns:

test statistics F

PValue

public double PValue()

Returns the p-Value of test statistics.

Returns:

p-Value

SampleSize

public int SampleSize()

Returns the sample size.

Returns:

n

NumIndepVar

public int NumIndepVar()

Returns the number of independent variables.

Returns:

k

toString

public java.lang.String toString()

Overrides:

toString in class java.lang.Object