gov.sandia.cognition.statistics.method
Class MultipleComparisonExperiment.Statistic

java.lang.Object
  extended by gov.sandia.cognition.util.AbstractCloneableSerializable
      extended by gov.sandia.cognition.statistics.method.MultipleComparisonExperiment.Statistic
All Implemented Interfaces:
ConfidenceStatistic, MultipleHypothesisComparison.Statistic, CloneableSerializable, Serializable, Cloneable
Enclosing class:
MultipleComparisonExperiment

public static class MultipleComparisonExperiment.Statistic
extends AbstractCloneableSerializable
implements ConfidenceStatistic, MultipleHypothesisComparison.Statistic

Result of running the MultipleHypothesisComparison hypothesis test

See Also:
Serialized Form

Field Summary
protected  ConfidenceStatistic blockExperimentResult
          Result from the block-experiment null-hypothesis test
protected  MultipleHypothesisComparison.Statistic multipleComparisonResult
          Result from the multiple hypothesis comparison null-hypothesis test, which will exist if the block-experiment null hypothesis is rejected
 
Constructor Summary
MultipleComparisonExperiment.Statistic(ConfidenceStatistic blockExperimentResult, MultipleHypothesisComparison.Statistic multipleComparisonResult)
          Creates a new instance of Statistic
 
Method Summary
 boolean acceptNullHypothesis(int i, int j)
          Determines if the (i,j) null hypothesis should be accepted (true) or rejected (false) .
 ConfidenceStatistic getBlockExperimentResult()
          Getter for blockExperimentResult
 MultipleHypothesisComparison.Statistic getMultipleComparisonResult()
          Getter for multipleComparisonResult
 double getNullHypothesisProbability()
          Gets the probability that the null hypothesis is correct.
 double getNullHypothesisProbability(int i, int j)
          Gets the Null Hypothesis probability associated with the (i,j) treatment comparison
 double getTestStatistic()
          Gets the statistic from which we compute the null-hypothesis probability.
 double getTestStatistic(int i, int j)
          Gets the test statistic associated with the (i,j) treatment comparison
 int getTreatmentCount()
          Gets the number of treatments being compared
 double getUncompensatedAlpha()
          Gets the uncompensated alpha (p-value threshold) for the multiple comparison test
 String toString()
           
 
Methods inherited from class gov.sandia.cognition.util.AbstractCloneableSerializable
clone
 
Methods inherited from class java.lang.Object
equals, finalize, getClass, hashCode, notify, notifyAll, wait, wait, wait
 
Methods inherited from interface gov.sandia.cognition.util.CloneableSerializable
clone
 

Field Detail

blockExperimentResult

protected ConfidenceStatistic blockExperimentResult
Result from the block-experiment null-hypothesis test


multipleComparisonResult

protected MultipleHypothesisComparison.Statistic multipleComparisonResult
Result from the multiple hypothesis comparison null-hypothesis test, which will exist if the block-experiment null hypothesis is rejected

Constructor Detail

MultipleComparisonExperiment.Statistic

public MultipleComparisonExperiment.Statistic(ConfidenceStatistic blockExperimentResult,
                                              MultipleHypothesisComparison.Statistic multipleComparisonResult)
Creates a new instance of Statistic

Parameters:
blockExperimentResult - Result from the block-experiment null-hypothesis test
multipleComparisonResult - Result from the multiple hypothesis comparison null-hypothesis test, which will exist if the block-experiment null hypothesis is rejected
Method Detail

getTestStatistic

public double getTestStatistic()
Description copied from interface: ConfidenceStatistic
Gets the statistic from which we compute the null-hypothesis probability. In an ANOVA, this would be the "F" statistic. In a t-test, this would be the "t" value. And so forth.

Specified by:
getTestStatistic in interface ConfidenceStatistic
Returns:
Confidence statistic used to compute the null-hypothesis probability.

getTreatmentCount

public int getTreatmentCount()
Description copied from interface: MultipleHypothesisComparison.Statistic
Gets the number of treatments being compared

Specified by:
getTreatmentCount in interface MultipleHypothesisComparison.Statistic
Returns:
Number of treatments being compared

getUncompensatedAlpha

public double getUncompensatedAlpha()
Description copied from interface: MultipleHypothesisComparison.Statistic
Gets the uncompensated alpha (p-value threshold) for the multiple comparison test

Specified by:
getUncompensatedAlpha in interface MultipleHypothesisComparison.Statistic
Returns:
Uncompensated alpha (p-value threshold) for the multiple comparison test

getTestStatistic

public double getTestStatistic(int i,
                               int j)
Description copied from interface: MultipleHypothesisComparison.Statistic
Gets the test statistic associated with the (i,j) treatment comparison

Specified by:
getTestStatistic in interface MultipleHypothesisComparison.Statistic
Parameters:
i - First treatment index
j - Second treatment index
Returns:
Test statistic associated with the (i,j) treatment comparison

getNullHypothesisProbability

public double getNullHypothesisProbability(int i,
                                           int j)
Description copied from interface: MultipleHypothesisComparison.Statistic
Gets the Null Hypothesis probability associated with the (i,j) treatment comparison

Specified by:
getNullHypothesisProbability in interface MultipleHypothesisComparison.Statistic
Parameters:
i - First treatment index
j - Second treatment index
Returns:
Null Hypothesis probability associated with the (i,j) treatment comparison

acceptNullHypothesis

public boolean acceptNullHypothesis(int i,
                                    int j)
Description copied from interface: MultipleHypothesisComparison.Statistic
Determines if the (i,j) null hypothesis should be accepted (true) or rejected (false) . Rejecting a null hypothesis typically means that there is a significant difference between the (i,j) treatment.

Specified by:
acceptNullHypothesis in interface MultipleHypothesisComparison.Statistic
Parameters:
i - First treatment index
j - Second treatment index
Returns:
True if we accept the null hypothesis, false if we reject the null hypothesis

getNullHypothesisProbability

public double getNullHypothesisProbability()
Description copied from interface: ConfidenceStatistic
Gets the probability that the null hypothesis is correct. That is, the probability that two datasets were generated by the same distribution. Generally, social scientists try to demonstrate that NullHypothesisProbability less than 0.05 to show that distributions are different. This is often called the "p-value" of a significance test.

Specified by:
getNullHypothesisProbability in interface ConfidenceStatistic
Returns:
Probability of the null hypothesis, often called "p-value"

getBlockExperimentResult

public ConfidenceStatistic getBlockExperimentResult()
Getter for blockExperimentResult

Returns:
Result from the block-experiment null-hypothesis test

getMultipleComparisonResult

public MultipleHypothesisComparison.Statistic getMultipleComparisonResult()
Getter for multipleComparisonResult

Returns:
Result from the multiple hypothesis comparison null-hypothesis test, which will exist if the block-experiment null hypothesis is rejected

toString

public String toString()
Overrides:
toString in class Object