gov.sandia.cognition.statistics.method

Class HolmCorrection

java.lang.Object

gov.sandia.cognition.util.AbstractCloneableSerializable

gov.sandia.cognition.statistics.method.AbstractMultipleHypothesisComparison<java.util.Collection<? extends java.lang.Number>,StatisticType>

gov.sandia.cognition.statistics.method.AbstractPairwiseMultipleHypothesisComparison<HolmCorrection.Statistic>

gov.sandia.cognition.statistics.method.HolmCorrection

All Implemented Interfaces:

MultipleHypothesisComparison<java.util.Collection<? extends java.lang.Number>>, CloneableSerializable, java.io.Serializable, java.lang.Cloneable

@PublicationReference(author="Wikipedia",
                      title="Holm\u2013Bonferroni method",
                      type=WebPage,
                      url="http://en.wikipedia.org/wiki/Holm%E2%80%93Bonferroni_method",
                      year=2011)
public class HolmCorrection
extends AbstractPairwiseMultipleHypothesisComparison<HolmCorrection.Statistic>

The Holm correction is a uniformly tighter bound than the Bonferroni/Sidak correction by first sorting the pair-wide p-values and then adjusting the p-values by the number of remaining hypotheses. To reject the first p-value, the smallest pair-wise p-value must be smaller than the Bonferroni corrected value (alpha/N), then the next smallest p-value must be smaller than (alpha/(N-1)) and so forth. This is a uniformly looser bound than the Shaffer static correction. However, the computational complexity of the Holm algorithm is quadratic in the number of treatments, meaning its linear in the number of actual comparisons. So, for treatments above 100 (4950 comparisons), the Holm correction may be the most appropriate choice. This implementation uses the slightly tighter Sidak correction, as opposed to the standard Bonferroni correction.

Since:

3.3.0

Author:

Kevin R. Dixon

See Also:

Serialized Form

Nested Class Summary

Nested Classes

Modifier and Type	Class and Description
static class	HolmCorrection.Statistic Test statistic from the Shaffer static multiple-comparison test

Field Summary

Fields inherited from class gov.sandia.cognition.statistics.method.AbstractPairwiseMultipleHypothesisComparison

DEFAULT_PAIRWISE_TEST, pairwiseTest

Fields inherited from interface gov.sandia.cognition.statistics.method.MultipleHypothesisComparison

DEFAULT_UNCOMPENSATED_ALPHA

Constructor Summary

Constructors

Constructor and Description
HolmCorrection() Default constructor
HolmCorrection(NullHypothesisEvaluator<java.util.Collection<? extends java.lang.Number>> pairwiseTest) Creates a new instance of BonferroniCorrection

Method Summary

Modifier and Type	Method and Description
HolmCorrection	clone() This makes public the clone method on the Object class and removes the exception that it throws.
HolmCorrection.Statistic	evaluateNullHypotheses(java.util.Collection<? extends java.util.Collection<? extends java.lang.Number>> data, double uncompensatedAlpha) Evaluates the null hypotheses associated with the given collection of data.

Methods inherited from class gov.sandia.cognition.statistics.method.AbstractPairwiseMultipleHypothesisComparison

getPairwiseTest, setPairwiseTest

Methods inherited from class gov.sandia.cognition.statistics.method.AbstractMultipleHypothesisComparison

evaluateNullHypotheses

Methods inherited from class java.lang.Object

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

Constructor Detail

HolmCorrection

public HolmCorrection()

Default constructor

HolmCorrection

public HolmCorrection(NullHypothesisEvaluator<java.util.Collection<? extends java.lang.Number>> pairwiseTest)

Creates a new instance of BonferroniCorrection

Parameters:

pairwiseTest - Confidence test used for pair-wise null-hypothesis tests.

Method Detail

clone

public HolmCorrection clone()

Description copied from class: AbstractCloneableSerializable

This makes public the clone method on the Object class and removes the exception that it throws. Its default behavior is to automatically create a clone of the exact type of object that the clone is called on and to copy all primitives but to keep all references, which means it is a shallow copy. Extensions of this class may want to override this method (but call super.clone() to implement a "smart copy". That is, to target the most common use case for creating a copy of the object. Because of the default behavior being a shallow copy, extending classes only need to handle fields that need to have a deeper copy (or those that need to be reset). Some of the methods in ObjectUtil may be helpful in implementing a custom clone method. Note: The contract of this method is that you must use super.clone() as the basis for your implementation.

Specified by:

clone in interface CloneableSerializable

Overrides:

clone in class AbstractPairwiseMultipleHypothesisComparison<HolmCorrection.Statistic>

Returns:

A clone of this object.

evaluateNullHypotheses

public HolmCorrection.Statistic evaluateNullHypotheses(java.util.Collection<? extends java.util.Collection<? extends java.lang.Number>> data,
                                                       double uncompensatedAlpha)

Description copied from interface: MultipleHypothesisComparison

Evaluates the null hypotheses associated with the given collection of data.

Specified by:

evaluateNullHypotheses in interface MultipleHypothesisComparison<java.util.Collection<? extends java.lang.Number>>

Specified by:

evaluateNullHypotheses in class AbstractMultipleHypothesisComparison<java.util.Collection<? extends java.lang.Number>,HolmCorrection.Statistic>

Parameters:

data - Data from each treatment to consider

uncompensatedAlpha - Uncompensated alpha (p-value threshold) for the multiple comparison test, must be [0,1]

Returns:

Statistic that summarizes the multiple comparison test