org.expasy.mzjava.stats

Class PoissonUtils

java.lang.Object

org.expasy.mzjava.stats.PoissonUtils

public class PoissonUtils
extends Object

Version:

sqrt -1

Author:

Oliver Horlacher

Method Summary

Methods

Modifier and Type	Method and Description
static double[]	confidenceInterval(int mean, double confidence) Calculates a confidence interval for a poisson distribution.
static double[]	confidenceInterval95(int mean) See http://www.math.mcmaster.ca/peter/s743/poissonalpha.html.
static double[]	hackRatioConfidenceInterval95(int m1, int m2)
static boolean	isSignificantlyDifferent(double i, double j, double significance)
static boolean	isSignificantlyDifferent(double x1, double y1, double x2, double y2, double significance)
static double[]	ratioConfidenceInterval95(int mean1, int mean2) Calculates the 95% confidence interval for the ratio of the two means.
static double	testSignificance(double x1, double x2) Retruns d for a test of significance between two Poisson counts in samples of equal counts.
static double	testSignificance(double x1, double y1, double x2, double y2) Retruns d for a test of significance between two Poisson counts where the sample sizes are unequal.

Methods inherited from class java.lang.Object

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

Method Detail

isSignificantlyDifferent

public static boolean isSignificantlyDifferent(double i,
                               double j,
                               double significance)

isSignificantlyDifferent

public static boolean isSignificantlyDifferent(double x1,
                               double y1,
                               double x2,
                               double y2,
                               double significance)

testSignificance

public static double testSignificance(double x1,
                      double x2)

Retruns d for a test of significance between two Poisson counts in samples of equal counts.

see http://books.google.co.nz/books?id=WovYdIgj6HMC&pg=PA34&lpg=PA34&dq=Comparison+of+two+Poisson+counts&source=bl&ots=eazoFoJUBV&sig=XIEtCDGLlXL7NzZdQws_lu1Iky4&hl=en&ei=uJh0Tv_yB8bwsgaB2NnBCw&sa=X&oi=book_result&ct=result&resnum=6&ved=0CFMQ6AEwBQ#v=onepage&q&f=false see http://www.biology.ed.ac.uk/research/groups/jdeacon/statistics/tress10.html#Poisson%20distribution%20for%20count%20data

Parameters:

x1 - mean of sample 1

x2 - mean of sample 2

Returns:

the d value

testSignificance

public static double testSignificance(double x1,
                      double y1,
                      double x2,
                      double y2)

Retruns d for a test of significance between two Poisson counts where the sample sizes are unequal.

see http://books.google.co.nz/books?id=WovYdIgj6HMC&pg=PA34&lpg=PA34&dq=Comparison+of+two+Poisson+counts&source=bl&ots=eazoFoJUBV&sig=XIEtCDGLlXL7NzZdQws_lu1Iky4&hl=en&ei=uJh0Tv_yB8bwsgaB2NnBCw&sa=X&oi=book_result&ct=result&resnum=6&ved=0CFMQ6AEwBQ#v=onepage&q&f=false see http://www.biology.ed.ac.uk/research/groups/jdeacon/statistics/tress10.html#Poisson%20distribution%20for%20count%20data

Parameters:

x1 - mean of sample 1

y1 - size of sample 1

x2 - mean of sample 2

y2 - size of sample 2

Returns:

the d value

ratioConfidenceInterval95

public static double[] ratioConfidenceInterval95(int mean1,
                                 int mean2)

Calculates the 95% confidence interval for the ratio of the two means.

Parameters:

mean1 - mean 1

mean2 - mean 2

Returns:

array containing the low and high values of the interval

hackRatioConfidenceInterval95

public static double[] hackRatioConfidenceInterval95(int m1,
                                     int m2)

confidenceInterval95

public static double[] confidenceInterval95(int mean)

See http://www.math.mcmaster.ca/peter/s743/poissonalpha.html.

confidenceInterval

public static double[] confidenceInterval(int mean,
                          double confidence)

Calculates a confidence interval for a poisson distribution.

Parameters:

mean - the mean

confidence - the confidence valid input is from 0 to 1

Returns:

the confidence interval