SuanShu, a Java numerical and statistical library



com.numericalmethod.suanshu.stats.test.variance
Class F

java.lang.Object
  extended by com.numericalmethod.suanshu.stats.test.HypothesisTest
      extended by com.numericalmethod.suanshu.stats.test.variance.F

public class F
extends HypothesisTest

The F-test tests whether two normal populations have the same variance. This test is sensitive to the assumption that the variables are normally distributed.

The R equivalent function is var.test.

See Also:
Wikipedia: FDistribution-test of equality of variances

Constructor Summary
F(double[] sample1, double[] sample2)
          Perform the F-test to test for equal variance of two normal populations.
F(double[] sample1, double[] sample2, double ratio)
          Perform the F-test to test for equal variance of two normal populations.
 
Method Summary
 double[] confidenceInterval(double confidence)
          Compute the confidence interval.
 String getAlternativeHypothesis()
          Get the description of the alternative hypothesis.
 String getNullHypothesis()
          Get a description of the null hypothesis.
 double leftConfidenceInterval(double confidence)
          Compute the one sided left confidence interval, [0, a]
 double leftOneSidedPvalue()
          Get the left, one-sided p-value.
 double pValue()
          Get the p-value for the test statistics.
 double rightConfidenceInterval(double confidence)
          Compute the one sided right confidence interval, [a, ∞)
 double rightOneSidedPvalue()
          Get the right, one-sided p-value.
 double statistics()
          Get the test statistics.
 
Methods inherited from class com.numericalmethod.suanshu.stats.test.HypothesisTest
isNullRejected, nGroups, nObs, oneSidedPvalue
 
Methods inherited from class java.lang.Object
clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait
 

Constructor Detail

F

public F(double[] sample1,
         double[] sample2)
Perform the F-test to test for equal variance of two normal populations.

Parameters:
sample1 - sample 1
sample2 - sample 2

F

public F(double[] sample1,
         double[] sample2,
         double ratio)
Perform the F-test to test for equal variance of two normal populations.

Parameters:
sample1 - sample 1
sample2 - sample 2
ratio - the hypothesized ratio of the population variances of samples 1 and 2
Method Detail

getNullHypothesis

public String getNullHypothesis()
Description copied from class: HypothesisTest
Get a description of the null hypothesis.

Specified by:
getNullHypothesis in class HypothesisTest
Returns:
the null hypothesis description
See Also:
Wikipedia: Null hypothesis

getAlternativeHypothesis

public String getAlternativeHypothesis()
Description copied from class: HypothesisTest
Get the description of the alternative hypothesis.

Specified by:
getAlternativeHypothesis in class HypothesisTest
Returns:
the alternative hypothesis description
See Also:
Wikipedia: Alternative hypothesis

statistics

public double statistics()
Description copied from class: HypothesisTest
Get the test statistics.

Specified by:
statistics in class HypothesisTest
Returns:
the test statistics
See Also:
Wikipedia: Test statistic

pValue

public double pValue()
Description copied from class: HypothesisTest
Get the p-value for the test statistics.

Specified by:
pValue in class HypothesisTest
Returns:
the p-value
See Also:
Wikipedia: P-value

rightOneSidedPvalue

public double rightOneSidedPvalue()
Get the right, one-sided p-value.

Returns:
the right, one-sided p-value.

leftOneSidedPvalue

public double leftOneSidedPvalue()
Get the left, one-sided p-value.

Returns:
the left, one-sided p-value.

confidenceInterval

public double[] confidenceInterval(double confidence)
Compute the confidence interval.

Parameters:
confidence - the confidence level, e.g., for a 2-sided 95% confidence interval, we use 0.975 because 1 - 0.95 = 2 * (1 - 0.025)
Returns:
the left and right interval bounds

rightConfidenceInterval

public double rightConfidenceInterval(double confidence)
Compute the one sided right confidence interval, [a, ∞)

Parameters:
confidence - the confidence level, e.g., 0.95 for 95% confidence interval
Returns:
the left interval bound

leftConfidenceInterval

public double leftConfidenceInterval(double confidence)
Compute the one sided left confidence interval, [0, a]

Parameters:
confidence - the confidence level, e.g., 0.95 for 95% confidence interval
Returns:
the right interval bound


Copyright © 2010-2014 Numerical Method Incorporation Limited. All Rights Reserved.