Package jasima.core.statistics

Class SamplingSummaryStat

java.lang.Object

jasima.core.statistics.SummaryStat

jasima.core.statistics.SamplingSummaryStat

All Implemented Interfaces:

Serializable, Cloneable

public class SamplingSummaryStat
extends SummaryStat

This class provides efficient sampling of up to a certain maximum number of values. A SummaryStat-object provides an efficient means to collect basic summary statistics like mean, min, max, standard deviation. To do so it does not have to store any observations. Obtaining information like median, or percentile requires storing values, however. To limit the amount of data which has to be stored, sampling can be used and the aforementioned statistics estimated based on this sample.

This class can sample and store up to a certain maximum number of values. Each value passed to a call of value(double) has the same probability of ending up in the final sample accessible by getData() .

Which values are selected is determined by a random number generator (see setRnd(Random)). If no such random number generator is set, Math.random() is used.

Author:

Torsten Hildebrandt, 2012-07-06

See Also:

Serialized Form

Nested Class Summary

Nested Classes

Modifier and Type	Class	Description
protected static class	SamplingSummaryStat.DatEntry	Each DatEntry stores a single decision situation plus a random value.

Field Summary

Fields inherited from class jasima.core.statistics.SummaryStat

DEF_ERROR_PROB, lastValue, lastWeight

Constructor Summary

Constructors

Constructor	Description
SamplingSummaryStat()
SamplingSummaryStat(int numSamples)
SamplingSummaryStat(int numSamples, Random rnd)
SamplingSummaryStat(Random rnd)

Method Summary

Modifier and Type	Method	Description
SummaryStat	clone()	Clones this object.
double[]	getData()
int	getNumSamples()
Random	getRnd()
void	init()
void	setNumSamples(int numSamples)
void	setRnd(Random rnd)
SamplingSummaryStat	value(double v)	Adds the given value with a weight of 1.
SummaryStat	value(double v, double weight)	Adds a value with a given weight.

Methods inherited from class jasima.core.statistics.SummaryStat

clear, combine, combine, confidenceIntervalLower, confidenceIntervalLower, confidenceIntervalUpper, confidenceIntervalUpper, confIntRangeSingle, lastValue, lastWeight, max, mean, min, numObs, stdDev, sum, summarize, summarize, values, varCoeff, variance, variancePopulation, weightSum

Methods inherited from class java.lang.Object

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

Constructor Detail

SamplingSummaryStat

public SamplingSummaryStat()

SamplingSummaryStat

public SamplingSummaryStat(int numSamples)

SamplingSummaryStat

public SamplingSummaryStat(Random rnd)

SamplingSummaryStat

public SamplingSummaryStat(int numSamples,
                           Random rnd)

Method Detail

value

public SummaryStat value(double v,
                         double weight)

Description copied from class: SummaryStat

Adds a value with a given weight.

Overrides:

value in class SummaryStat

Parameters:

v - The value to add.

weight - The weight to give to this value. Has to be positive.

Returns:

this, to allow easy chaining of calls.

value

public SamplingSummaryStat value(double v)

Description copied from class: SummaryStat

Adds the given value with a weight of 1.

Overrides:

value in class SummaryStat

Parameters:

v - The value to add.

Returns:

this, to allow easy chaining of calls.

getData

public double[] getData()

init

public void init()

setRnd

public void setRnd(Random rnd)

getRnd

public Random getRnd()

setNumSamples

public void setNumSamples(int numSamples)

getNumSamples

public int getNumSamples()

clone

public SummaryStat clone()

Description copied from class: SummaryStat

Clones this object. We can use the standard functionality here, as there are only primitive fields.

Overrides:

clone in class SummaryStat

Returns:

A clone of this SummaryStat.