Package com.bayesserver.analysis

Class HistogramDensity

java.lang.Object

com.bayesserver.analysis.HistogramDensity

All Implemented Interfaces:

EmpiricalDensity

public final class HistogramDensity
extends Object
implements EmpiricalDensity

Represents an empirical density function built from a histogram, which can represent arbitrary univariate distributions.

Constructor Summary

Constructors

Constructor	Description
HistogramDensity(List<Interval<Double>> intervals, List<Double> intervalCounts)	Constructs an empirical density function.

Method Summary

Modifier and Type	Method	Description
double	cdf(double x)	Calculates an approximate value for cdf(x).
List<HistogramDensityItem>	getItems()	The collection of intervals and their statistics making up the histogram density.
double	inverseCdf(double probability)	Calculates an approximate value for the inverse cumulative distribution function.
static HistogramDensity	learn(Iterable<WeightedValue> values, HistogramDensityOptions options)	Learns a univariate empirical density from data.

Methods inherited from class java.lang.Object

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

Constructor Detail

HistogramDensity

public HistogramDensity(List<Interval<Double>> intervals,
                        List<Double> intervalCounts)

Constructs an empirical density function.

Parameters:

intervals - The intervals making up the approximate density.

intervalCounts - The number of elements in each interval.

Method Detail

inverseCdf

public double inverseCdf(double probability)

Calculates an approximate value for the inverse cumulative distribution function.

Specified by:

inverseCdf in interface EmpiricalDensity

Parameters:

probability - The probability p at which to return x when p = Cdf(x) .

Returns:

The approximate inverse cdf.

getItems

public List<HistogramDensityItem> getItems()

The collection of intervals and their statistics making up the histogram density.

cdf

public double cdf(double x)

Calculates an approximate value for cdf(x).

Specified by:

cdf in interface EmpiricalDensity

Parameters:

x - The value at which to evaluate the cdf.

Returns:

The approximate cdf(x).

learn

public static HistogramDensity learn(Iterable<WeightedValue> values,
                                     HistogramDensityOptions options)

Learns a univariate empirical density from data.

Parameters:

values - The values to learn from.

options - Options affecting the learning process.

Returns:

An empirical density.