pamMaths
Class PamHistogram

java.lang.Object
  java.util.Observable
      PamUtils.TimedObservable
          pamMaths.PamHistogram

Direct Known Subclasses:

TriggerHistogram

public class PamHistogram
extends TimedObservable

class for collecting data into a 1-D histogram.

As well as basic histogramming functions, functions exist for downscaling the histogram data and extracting parameters such as variance, skew, etc.

Author:

Doug Gillespie

Field Summary

private boolean	binCentres
protected double[]	data
protected double	hiBin
private double	histoScale
protected double	loBin
private double	maxVal
private double	minVal
private java.lang.String	name
private int	nBins

Constructor Summary

PamHistogram(double minVal, double maxVal, int nBins)
Constructs a histogram object with minVal and maxVal, specifying the low edge of the lowest bin and the high edge of the highest bin respectively.

PamHistogram(double minVal, double maxVal, int nBins, boolean binCentres)
Constructs a histogram object where minVal and maxVal specify the centres of the lower and upper bins (true) or the lower edge of the lowest bin and the upper edge of the highest bin (false)

Method Summary

void	addData(double newData) Add a single point of data to the histogram but don't notify observers
void	addData(double newData, boolean notify) Add a single unit value to the histogram and optionally notify all observers
void	addData(double newData, double weight) Add weighted data to the histogram
void	clear() Clear all histogram contents
int	getBin(double dataValue) Work out which bin a particular value will call into.
double	getBinCentre(int iBin)
double[]	getData() GEt the histograms data array.
double	getHiBin()
double	getKurtosis() Get the kurtosis of the distribution (fourth moment) taking the definition from Numerical recipes in C p 612
double	getLoBin()
double	getMaxContent() Get max bin content, including lo and high fields.
double	getMaxVal() Get the maximum range value of the histogram
double	getMean() Calculate the mean of the histogam data
double	getMinVal() Get the minimum range value of the histogram
double	getMode() Get's the modal value of the distribution
java.lang.String	getName()
int	getNBins() Get the number of histogram bins
double	getScaleMaxVal() Where to plot to if bin centres were specified.
double	getScaleMinVal() Where to plot from if bin centres were specified.
double	getSkew() Get the skewness of the distribution (third moment)
double	getSTD() Return the standard deviation of the histogram data
double	getStep() Step size between sucessive bins.
double	getTotalContent() Get's the sum total of all bin contents, excluding lo and high bins
double	getVariance() Return the variance of the histogram data
boolean	isBinCentres()
private void	prepareHistogram(double minVal, double maxVal, int nBins, boolean binCentres) Allocates memory and generally prepeare the histogram for use.
void	scaleData(double scaleFactor) Scale the data in the histogram by the given factor
void	setBinCentres(boolean binCentres)
void	setName(java.lang.String name)
void	setRange(double minVal, double maxVal, int nBins) Sets the histogram bin ranges.
java.lang.String	toString()

Methods inherited from class PamUtils.TimedObservable

getDelay, notifyObservers, notifyObservers, notifyObservers, setDelay

Methods inherited from class java.util.Observable

addObserver, clearChanged, countObservers, deleteObserver, deleteObservers, hasChanged, setChanged

Methods inherited from class java.lang.Object

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

Field Detail

data

protected double[] data

loBin

protected double loBin

hiBin

protected double hiBin

minVal

private double minVal

maxVal

private double maxVal

nBins

private int nBins

histoScale

private double histoScale

binCentres

private boolean binCentres

name

private java.lang.String name

Constructor Detail

PamHistogram

public PamHistogram(double minVal,
                    double maxVal,
                    int nBins)

Constructs a histogram object with minVal and maxVal, specifying the low edge of the lowest bin and the high edge of the highest bin respectively.

Parameters:

minVal - min Value

maxVal - max Value

nBins - number of bins

PamHistogram

public PamHistogram(double minVal,
                    double maxVal,
                    int nBins,
                    boolean binCentres)

Constructs a histogram object where minVal and maxVal specify the centres of the lower and upper bins (true) or the lower edge of the lowest bin and the upper edge of the highest bin (false)

Parameters:

minVal - minimum value (low edge or low bin centre, depending on binCentres parameter)

maxVal - maximum value (high edge or high bin centre, depending on binCentres parameter)

nBins - total number of bins

binCentres - specifies whether the minVal and maxVal specify the centres of the lower and upper bins (true) or the lower edge of the lowest bin and the upper edge of the highest bin (false)

Method Detail

prepareHistogram

private void prepareHistogram(double minVal,
                              double maxVal,
                              int nBins,
                              boolean binCentres)

Allocates memory and generally prepeare the histogram for use.

Parameters:

minVal - minimum value (low edge or low bin centre, depending on binCentres parameter)

maxVal - maximum value (high edge or high bin centre, depending on binCentres parameter)

nBins - total number of bins

binCentres - specifies whether the minVal and maxVal specify the centres of the lower and upper bins (true) or the lower edge of the lowest bin and the upper edge of the highest bin (false)

setRange

public void setRange(double minVal,
                     double maxVal,
                     int nBins)

Sets the histogram bin ranges.

Note that calling this function will reset the histogram data.

Parameters:

minVal - minimum value (low edge or low bin centre, depending on binCentres parameter)

maxVal - maximum value (high edge or high bin centre, depending on binCentres parameter)

nBins - total number of bins

getBin

public int getBin(double dataValue)

Work out which bin a particular value will call into. N.B. This may return < 0 or >= getNBins() if the data fall outsied the range of the histogram

Parameters:

dataValue -

Returns:

bin Bumber

getBinCentre

public double getBinCentre(int iBin)

addData

public void addData(double newData,
                    boolean notify)

Add a single unit value to the histogram and optionally notify all observers

Parameters:

newData - value of the new data

addData

public void addData(double newData)

Add a single point of data to the histogram but don't notify observers

Parameters:

newData - new data point

addData

public void addData(double newData,
                    double weight)

Add weighted data to the histogram

Parameters:

newData - value of the data

weight - weight to apply

clear

public void clear()

Clear all histogram contents

scaleData

public void scaleData(double scaleFactor)

Scale the data in the histogram by the given factor

This is used in various parts of PAMGUARD which wish to keep a decaying average distribution - every few seconds the data are halved (or some such) so that all data are represented, but recent data will carry more weight in the distributions.

Parameters:

scaleFactor - FActor to scale data by (generally < 1)

getMaxContent

public double getMaxContent()

Get max bin content, including lo and high fields.

Returns:

max bin content.

getMean

public double getMean()

Calculate the mean of the histogam data

Returns:

mean value

getSTD

public double getSTD()

Return the standard deviation of the histogram data

Returns:

standard deviation

getVariance

public double getVariance()

Return the variance of the histogram data

Returns:

get variance of distribution in histogram

getSkew

public double getSkew()

Get the skewness of the distribution (third moment)

Returns:

skew

getKurtosis

public double getKurtosis()

Get the kurtosis of the distribution (fourth moment) taking the definition from Numerical recipes in C p 612

Returns:

kurtosis

getMode

public double getMode()

Get's the modal value of the distribution

Returns:

value of the bin centre for the most populated bin

getTotalContent

public double getTotalContent()

Get's the sum total of all bin contents, excluding lo and high bins

Returns:

get total content of all histogram bins

getData

public double[] getData()

GEt the histograms data array.

Returns:

An array of double data

getMaxVal

public double getMaxVal()

Get the maximum range value of the histogram

Returns:

max value

getMinVal

public double getMinVal()

Get the minimum range value of the histogram

Returns:

min value

getScaleMinVal

public double getScaleMinVal()

Where to plot from if bin centres were specified.

Returns:

minimum value for plot axis

getScaleMaxVal

public double getScaleMaxVal()

Where to plot to if bin centres were specified.

Returns:

maximum value for plot axis

getNBins

public int getNBins()

Get the number of histogram bins

Returns:

number of histogram bins

getStep

public double getStep()

Step size between sucessive bins.

Returns:

histogram step size (bin width)

getName

public java.lang.String getName()

setName

public void setName(java.lang.String name)

toString

public java.lang.String toString()

Overrides:

toString in class java.lang.Object

getHiBin

public double getHiBin()

getLoBin

public double getLoBin()

isBinCentres

public boolean isBinCentres()

setBinCentres

public void setBinCentres(boolean binCentres)