de.lmu.ifi.dbs.elki.math

Class MeanVariance

java.lang.Object

de.lmu.ifi.dbs.elki.math.Mean

de.lmu.ifi.dbs.elki.math.MeanVariance

Direct Known Subclasses:

MeanVarianceMinMax

@Reference(authors="B. P. Welford",
           title="Note on a method for calculating corrected sums of squares and products",
           booktitle="Technometrics 4(3)")
public class MeanVariance
extends Mean

Do some simple statistics (mean, variance) using a numerically stable online algorithm. This class can repeatedly be fed with data using the add() methods, the resulting values for mean and average can be queried at any time using getMean() and getSampleVariance(). Make sure you have understood variance correctly when using getNaiveVariance() - since this class is fed with samples and estimates the mean from the samples, getSampleVariance() is the proper formula. Trivial code, but replicated a lot. The class is final so it should come at low cost. Related Literature:

B. P. Welford
Note on a method for calculating corrected sums of squares and products
in: Technometrics 4(3)

D.H.D. West
Updating Mean and Variance Estimates: An Improved Method
In: Communications of the ACM, Volume 22 Issue 9

Author:

Erich Schubert

Field Summary

Fields

Modifier and Type	Field and Description
protected double	m2 nVariance

Fields inherited from class de.lmu.ifi.dbs.elki.math.Mean

m1, n

Constructor Summary

Constructors

Constructor and Description
MeanVariance() Empty constructor
MeanVariance(MeanVariance other) Constructor from other instance

Method Summary

Methods

Modifier and Type	Method and Description
double	denormalizeValue(double val) Return the unnormalized value (centered at the mean, distance normalized by standard deviation)
double	getCount() Get the number of points the average is based on.
double	getMean() Return mean
double	getNaiveStddev() Return standard deviation using the non-sample variance Note: usually, you should be using getSampleStddev() instead!
double	getNaiveVariance() Return the naive variance (not taking sampling into account) Note: usually, you should be using getSampleVariance() instead!
double	getSampleStddev() Return standard deviation
double	getSampleVariance() Return sample variance.
static MeanVariance[]	newArray(int dimensionality) Create and initialize a new array of MeanVariance
double	normalizeValue(double val) Return the normalized value (centered at the mean, distance normalized by standard deviation)
void	put(double val) Add a single value with weight 1.0
MeanVariance	put(double[] vals) Add values with weight 1.0
MeanVariance	put(double[] vals, double[] weights) Add values with weight 1.0
void	put(double val, double weight) Add data with a given weight.
void	put(Mean other) Join the data of another MeanVariance instance.
void	reset() Reset the value.
String	toString()

Methods inherited from class de.lmu.ifi.dbs.elki.math.Mean

of

Methods inherited from class java.lang.Object

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

Field Detail

m2

protected double m2

nVariance

Constructor Detail

MeanVariance

public MeanVariance()

Empty constructor

MeanVariance

public MeanVariance(MeanVariance other)

Constructor from other instance

Parameters:

other - other instance to copy data from.

Method Detail

put

public void put(double val)

Add a single value with weight 1.0

Overrides:

put in class Mean

Parameters:

val - Value

put

@Reference(authors="D.H.D. West",
           title="Updating Mean and Variance Estimates: An Improved Method",
           booktitle="Communications of the ACM, Volume 22 Issue 9")
public void put(double val,
                                      double weight)

Add data with a given weight. See also: D.H.D. West
Updating Mean and Variance Estimates: An Improved Method

Overrides:

put in class Mean

Parameters:

val - data

weight - weight

put

public void put(Mean other)

Join the data of another MeanVariance instance.

Overrides:

put in class Mean

Parameters:

other - Data to join with

put

public MeanVariance put(double[] vals)

Add values with weight 1.0

Overrides:

put in class Mean

Parameters:

vals - Values

Returns:

this

put

public MeanVariance put(double[] vals,
               double[] weights)

Description copied from class: Mean

Add values with weight 1.0

Overrides:

put in class Mean

Parameters:

vals - Values

Returns:

this

getCount

public double getCount()

Get the number of points the average is based on.

Overrides:

getCount in class Mean

Returns:

number of data points

getMean

public double getMean()

Return mean

Overrides:

getMean in class Mean

Returns:

mean

getNaiveVariance

public double getNaiveVariance()

Return the naive variance (not taking sampling into account) Note: usually, you should be using getSampleVariance() instead!

Returns:

variance

getSampleVariance

public double getSampleVariance()

Return sample variance.

Returns:

sample variance

getNaiveStddev

public double getNaiveStddev()

Return standard deviation using the non-sample variance Note: usually, you should be using getSampleStddev() instead!

Returns:

stddev

getSampleStddev

public double getSampleStddev()

Return standard deviation

Returns:

stddev

normalizeValue

public double normalizeValue(double val)

Return the normalized value (centered at the mean, distance normalized by standard deviation)

Parameters:

val - original value

Returns:

normalized value

denormalizeValue

public double denormalizeValue(double val)

Return the unnormalized value (centered at the mean, distance normalized by standard deviation)

Parameters:

val - normalized value

Returns:

de-normalized value

newArray

public static MeanVariance[] newArray(int dimensionality)

Create and initialize a new array of MeanVariance

Parameters:

dimensionality - Dimensionality

Returns:

New and initialized Array

toString

public String toString()

Overrides:

toString in class Mean

reset

public void reset()

Description copied from class: Mean

Reset the value.

Overrides:

reset in class Mean