Package org.meteoinfo.math.stats

Class StatsUtil

java.lang.Object

org.meteoinfo.math.stats.StatsUtil

public class StatsUtil
extends java.lang.Object

Constructor Summary

Constructors

Constructor	Description
StatsUtil()

Method Summary

Modifier and Type	Method	Description
static double[]	chiSquareTest(ucar.ma2.Array o)	Chi-square test of independence
static double[]	chiSquareTest(ucar.ma2.Array e, ucar.ma2.Array o)	Chi-square test
static java.lang.Object	cov(ucar.ma2.Array a, boolean bias)	Computes covariances for columns of a matrix.
static ucar.ma2.Array	cov(ucar.ma2.Array x, ucar.ma2.Array y, boolean bias)	Computes covariances for pairs of arrays or columns of a matrix.
static double	covariance(ucar.ma2.Array x, ucar.ma2.Array y, boolean bias)	Computes covariance of two arrays.
static double	kendalltau(ucar.ma2.Array x, ucar.ma2.Array y)	Calculates Kendall's tau, a correlation measure for ordinal data.
static ucar.ma2.Array[]	multipleLineRegress_OLS(ucar.ma2.Array y, ucar.ma2.Array x)	Implements ordinary least squares (OLS) to estimate the parameters of a multiple linear regression model.
static ucar.ma2.Array[]	multipleLineRegress_OLS(ucar.ma2.Array y, ucar.ma2.Array x, boolean noIntercept)	Implements ordinary least squares (OLS) to estimate the parameters of a multiple linear regression model.
static double[]	pairedTTest(ucar.ma2.Array a, ucar.ma2.Array b)	Paired test evaluating the null hypothesis that the mean difference between corresponding (paired) elements of the double[] arrays sample1 and sample2 is zero.
static double[]	pearsonr(ucar.ma2.Array x, ucar.ma2.Array y)	Calculates a Pearson correlation coefficient.
static double	percentile(ucar.ma2.Array a, double p)	Returns an estimate of the pth percentile of the values in the array.
static ucar.ma2.Array	percentile(ucar.ma2.Array a, double p, int axis)	Returns an estimate of the pth percentile of the values in the array along an axis.
static java.lang.Object	spearmanr(ucar.ma2.Array a)	Computes Spearman's rank correlation for columns of a matrix.
static ucar.ma2.Array	spearmanr(ucar.ma2.Array x, ucar.ma2.Array y)	Computes Spearman's rank correlation for pairs of arrays or columns of a matrix.
static double[]	tTest(ucar.ma2.Array a, double mu)	One sample t test
static double[]	tTest(ucar.ma2.Array a, ucar.ma2.Array b)	unpaired, two-sided, two-sample t-test.

Methods inherited from class java.lang.Object

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

Constructor Detail

StatsUtil

public StatsUtil()

Method Detail

covariance

public static double covariance(ucar.ma2.Array x,
                                ucar.ma2.Array y,
                                boolean bias)

Computes covariance of two arrays.

Parameters:

x - X data

y - Y data

bias - If true, returned value will be bias-corrected

Returns:

The covariance

cov

public static ucar.ma2.Array cov(ucar.ma2.Array x,
                                 ucar.ma2.Array y,
                                 boolean bias)

Computes covariances for pairs of arrays or columns of a matrix.

Parameters:

x - X data

y - Y data

bias - If true, returned value will be bias-corrected

Returns:

The covariance matrix

cov

public static java.lang.Object cov(ucar.ma2.Array a,
                                   boolean bias)

Computes covariances for columns of a matrix.

Parameters:

a - Matrix data

bias - If true, returned value will be bias-corrected

Returns:

Covariant matrix or value

kendalltau

public static double kendalltau(ucar.ma2.Array x,
                                ucar.ma2.Array y)

Calculates Kendall's tau, a correlation measure for ordinal data.

Parameters:

x - X data

y - Y data

Returns:

Kendall's tau correlation.

pearsonr

public static double[] pearsonr(ucar.ma2.Array x,
                                ucar.ma2.Array y)

Calculates a Pearson correlation coefficient.

Parameters:

x - X data

y - Y data

Returns:

Pearson correlation and p-value.

spearmanr

public static ucar.ma2.Array spearmanr(ucar.ma2.Array x,
                                       ucar.ma2.Array y)

Computes Spearman's rank correlation for pairs of arrays or columns of a matrix.

Parameters:

x - X data

y - Y data

Returns:

Spearman's rank correlation

spearmanr

public static java.lang.Object spearmanr(ucar.ma2.Array a)

Computes Spearman's rank correlation for columns of a matrix.

Parameters:

a - Matrix data

Returns:

Spearman's rank correlation

multipleLineRegress_OLS

public static ucar.ma2.Array[] multipleLineRegress_OLS(ucar.ma2.Array y,
                                                       ucar.ma2.Array x)

Implements ordinary least squares (OLS) to estimate the parameters of a multiple linear regression model.

Parameters:

y - Y sample data - one dimension array

x - X sample data - two dimension array

Returns:

Estimated regression parameters and residuals

multipleLineRegress_OLS

public static ucar.ma2.Array[] multipleLineRegress_OLS(ucar.ma2.Array y,
                                                       ucar.ma2.Array x,
                                                       boolean noIntercept)

Implements ordinary least squares (OLS) to estimate the parameters of a multiple linear regression model.

Parameters:

y - Y sample data - one dimension array

x - X sample data - two dimension array

noIntercept - No intercept

Returns:

Estimated regression parameters and residuals

percentile

public static double percentile(ucar.ma2.Array a,
                                double p)

Returns an estimate of the pth percentile of the values in the array.

Parameters:

a - Input array

p - The percentile value to compute

Returns:

The pth percentile

percentile

public static ucar.ma2.Array percentile(ucar.ma2.Array a,
                                        double p,
                                        int axis)
                                 throws ucar.ma2.InvalidRangeException

Returns an estimate of the pth percentile of the values in the array along an axis.

Parameters:

a - Input array

p - The percentile value to compute

axis - The axis

Returns:

The pth percentile

Throws:

ucar.ma2.InvalidRangeException

tTest

public static double[] tTest(ucar.ma2.Array a,
                             double mu)

One sample t test

Parameters:

a - Input data

mu - Expected value in null hypothesis

Returns:

t_statistic and p_value

tTest

public static double[] tTest(ucar.ma2.Array a,
                             ucar.ma2.Array b)

unpaired, two-sided, two-sample t-test.

Parameters:

a - Sample a.

b - Sample b.

Returns:

t_statistic and p_value

pairedTTest

public static double[] pairedTTest(ucar.ma2.Array a,
                                   ucar.ma2.Array b)

Paired test evaluating the null hypothesis that the mean difference between corresponding (paired) elements of the double[] arrays sample1 and sample2 is zero.

Parameters:

a - Sample a.

b - Sample b.

Returns:

t_statistic and p_value

chiSquareTest

public static double[] chiSquareTest(ucar.ma2.Array e,
                                     ucar.ma2.Array o)

Chi-square test

Parameters:

e - Expected.

o - Observed.

Returns:

Chi-square_statistic and p_value

chiSquareTest

public static double[] chiSquareTest(ucar.ma2.Array o)

Chi-square test of independence

Parameters:

o - Observed.

Returns:

Chi-square_statistic and p_value