org.hipparchus.stat.inference

Class OneWayAnova

java.lang.Object

org.hipparchus.stat.inference.OneWayAnova

public class OneWayAnova
extends Object

Implements one-way ANOVA (analysis of variance) statistics.

Tests for differences between two or more categories of univariate data (for example, the body mass index of accountants, lawyers, doctors and computer programmers). When two categories are given, this is equivalent to the TTest.

Uses the Hipparchus F Distribution implementation to estimate exact p-values.

This implementation is based on a description at http://faculty.vassar.edu/lowry/ch13pt1.html

 Abbreviations: bg = between groups,
                wg = within groups,
                ss = sum squared deviations
 

Constructor Summary

Constructors

Constructor and Description
OneWayAnova()

Method Summary

Modifier and Type	Method and Description
double	anovaFValue(Collection<double[]> categoryData) Computes the ANOVA F-value for a collection of double[] arrays.
double	anovaPValue(Collection<double[]> categoryData) Computes the ANOVA P-value for a collection of double[] arrays.
double	anovaPValue(Collection<StreamingStatistics> categoryData, boolean allowOneElementData) Computes the ANOVA P-value for a collection of StreamingStatistics.
boolean	anovaTest(Collection<double[]> categoryData, double alpha) Performs an ANOVA test, evaluating the null hypothesis that there is no difference among the means of the data categories.

Methods inherited from class java.lang.Object

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

Constructor Detail

OneWayAnova

public OneWayAnova()

Method Detail

anovaFValue

public double anovaFValue(Collection<double[]> categoryData)
                   throws MathIllegalArgumentException,
                          NullArgumentException

Computes the ANOVA F-value for a collection of double[] arrays.

Preconditions:

• The categoryData Collection must contain double[] arrays.
• There must be at least two double[] arrays in the categoryData collection and each of these arrays must contain at least two values.

This implementation computes the F statistic using the definitional formula

   F = msbg/mswg

where

  msbg = between group mean square
  mswg = within group mean square

are as defined here

Parameters:

categoryData - Collection of double[] arrays each containing data for one category

Returns:

Fvalue

Throws:

NullArgumentException - if categoryData is null

MathIllegalArgumentException - if the length of the categoryData array is less than 2 or a contained double[] array does not have at least two values

anovaPValue

public double anovaPValue(Collection<double[]> categoryData)
                   throws MathIllegalArgumentException,
                          NullArgumentException,
                          MathIllegalStateException

Computes the ANOVA P-value for a collection of double[] arrays.

Preconditions:

• The categoryData Collection must contain double[] arrays.
• There must be at least two double[] arrays in the categoryData collection and each of these arrays must contain at least two values.

This implementation uses the Hipparchus F Distribution implementation to estimate the exact p-value, using the formula

   p = 1 - cumulativeProbability(F)

where F is the F value and cumulativeProbability is the Hipparchus implementation of the F distribution.

Parameters:

categoryData - Collection of double[] arrays each containing data for one category

Returns:

Pvalue

Throws:

NullArgumentException - if categoryData is null

MathIllegalArgumentException - if the length of the categoryData array is less than 2 or a contained double[] array does not have at least two values

MathIllegalStateException - if the p-value can not be computed due to a convergence error

MathIllegalStateException - if the maximum number of iterations is exceeded

anovaPValue

public double anovaPValue(Collection<StreamingStatistics> categoryData,
                          boolean allowOneElementData)
                   throws MathIllegalArgumentException,
                          NullArgumentException,
                          MathIllegalStateException

Computes the ANOVA P-value for a collection of StreamingStatistics.

Preconditions:

• The categoryData Collection must contain StreamingStatistics.
• There must be at least two StreamingStatistics in the categoryData collection and each of these statistics must contain at least two values.

This implementation uses the Hipparchus F Distribution implementation to estimate the exact p-value, using the formula

   p = 1 - cumulativeProbability(F)

where F is the F value and cumulativeProbability is the Hipparchus implementation of the F distribution.

Parameters:

categoryData - Collection of StreamingStatistics each containing data for one category

allowOneElementData - if true, allow computation for one catagory only or for one data element per category

Returns:

Pvalue

Throws:

NullArgumentException - if categoryData is null

MathIllegalArgumentException - if the length of the categoryData array is less than 2 or a contained StreamingStatistics does not have at least two values

MathIllegalStateException - if the p-value can not be computed due to a convergence error

MathIllegalStateException - if the maximum number of iterations is exceeded

anovaTest

public boolean anovaTest(Collection<double[]> categoryData,
                         double alpha)
                  throws MathIllegalArgumentException,
                         NullArgumentException,
                         MathIllegalStateException

Performs an ANOVA test, evaluating the null hypothesis that there is no difference among the means of the data categories.

Preconditions:

• The categoryData Collection must contain double[] arrays.
• There must be at least two double[] arrays in the categoryData collection and each of these arrays must contain at least two values.
• alpha must be strictly greater than 0 and less than or equal to 0.5.

This implementation uses the Hipparchus F Distribution implementation to estimate the exact p-value, using the formula

   p = 1 - cumulativeProbability(F)

where F is the F value and cumulativeProbability is the Hipparchus implementation of the F distribution.

True is returned iff the estimated p-value is less than alpha.

Parameters:

categoryData - Collection of double[] arrays each containing data for one category

alpha - significance level of the test

Returns:

true if the null hypothesis can be rejected with confidence 1 - alpha

Throws:

NullArgumentException - if categoryData is null

MathIllegalArgumentException - if the length of the categoryData array is less than 2 or a contained double[] array does not have at least two values

MathIllegalArgumentException - if alpha is not in the range (0, 0.5]

MathIllegalStateException - if the p-value can not be computed due to a convergence error

MathIllegalStateException - if the maximum number of iterations is exceeded