gov.sandia.cognition.statistics.distribution

Class ScalarMixtureDensityModel

java.lang.Object

gov.sandia.cognition.util.AbstractCloneableSerializable

gov.sandia.cognition.statistics.AbstractDistribution<DataType>

gov.sandia.cognition.statistics.distribution.LinearMixtureModel<java.lang.Double,SmoothUnivariateDistribution>

gov.sandia.cognition.statistics.distribution.ScalarMixtureDensityModel

All Implemented Interfaces:

Vectorizable, ClosedFormComputableDistribution<java.lang.Double>, ClosedFormDistribution<java.lang.Double>, ClosedFormUnivariateDistribution<java.lang.Double>, ComputableDistribution<java.lang.Double>, Distribution<java.lang.Double>, DistributionWithMean<java.lang.Double>, SmoothUnivariateDistribution, UnivariateDistribution<java.lang.Double>, CloneableSerializable, java.io.Serializable, java.lang.Cloneable

Direct Known Subclasses:

ScalarMixtureDensityModel.CDF, ScalarMixtureDensityModel.PDF

@CodeReview(reviewer="Kevin R. Dixon",
            date="2009-10-20",
            changesNeeded=true,
            comments={"Fixed some missing javadoc.","General style fixes.","Added task to figure out a way to avoid storing weights in matrix.","Generally looks good.","Some argument checks need to be more complete"},
            response=@CodeReviewResponse(date="2009-10-20",respondent="Dan Morrow",comments={"added additional test coverage","added more argument checks"},moreChangesNeeded=false))
 @PublicationReference(author="Wikipedia",
                      title="Mixture Model",
                      type=WebPage,
                      year=2009,
                      url="http://en.wikipedia.org/wiki/Mixture_density")
public class ScalarMixtureDensityModel
extends LinearMixtureModel<java.lang.Double,SmoothUnivariateDistribution>
implements SmoothUnivariateDistribution

ScalarMixtureDensityModel (SMDM) implements just that: a scalar mixture density model. There are n distributions which can each be different. There is an n-dimensional vector of prior probabilities which are the probability of selecting each particular distribution. So these prior probabilities must sum to 1.0. To sample from a SMDM is to first select which distribution using the prior probabilities, and then to sample from that distribution to return a sample.

Each distribution must have a mean and variance defined. A mean and variance for the SMDM can be computed. Given an input value, a weighted Z value can be computed for the SMDM distribution.

Author:

jdmorr

See Also:

Serialized Form

Nested Class Summary

Nested Classes

Modifier and Type	Class and Description
static class	ScalarMixtureDensityModel.CDF CDFof the SMDM
static class	ScalarMixtureDensityModel.EMLearner An EM learner that estimates a mixture model from data
static class	ScalarMixtureDensityModel.PDF PDF of the SMDM

Field Summary

Fields inherited from class gov.sandia.cognition.statistics.distribution.LinearMixtureModel

distributions, priorWeights

Constructor Summary

Constructors

Constructor and Description
ScalarMixtureDensityModel() Creates a new instance of ScalarMixtureDensityModel
ScalarMixtureDensityModel(java.util.Collection<? extends SmoothUnivariateDistribution> distributions) Creates a new instance of ScalarMixtureDensityModel
ScalarMixtureDensityModel(java.util.Collection<? extends SmoothUnivariateDistribution> distributions, double[] priorWeights) Creates a new instance of ScalarMixtureDensityModel
ScalarMixtureDensityModel(ScalarMixtureDensityModel other) Copy constructor
ScalarMixtureDensityModel(SmoothUnivariateDistribution... distributions) Creates a new instance of ScalarMixtureDensityModel

Method Summary

Modifier and Type	Method and Description
ScalarMixtureDensityModel	clone() This makes public the clone method on the Object class and removes the exception that it throws.
void	convertFromVector(Vector parameters) Converts the object from a Vector of parameters.
Vector	convertToVector() Converts the object to a vector.
ScalarMixtureDensityModel.CDF	getCDF() Gets the CDF of a scalar distribution.
java.lang.Double	getMaxSupport() Gets the minimum support (domain or input) of the distribution.
java.lang.Double	getMean() Gets the arithmetic mean, or "first central moment" or "expectation", of the distribution.
double	getMeanAsDouble() Gets the mean of the distribution as a double.
java.lang.Double	getMinSupport() Gets the minimum support (domain or input) of the distribution.
ScalarMixtureDensityModel.PDF	getProbabilityFunction() Gets the distribution function associated with this Distribution, either the PDF or PMF.
double	getVariance() Gets the variance of the distribution.
double	sampleAsDouble(java.util.Random random) Samples a value from this distribution as a double.
double[]	sampleAsDoubles(java.util.Random random, int count) Samples values from this distribution as an array of doubles.
void	sampleInto(java.util.Random random, double[] output, int start, int length) Samples values from this distribution as an array of doubles.

Methods inherited from class gov.sandia.cognition.statistics.distribution.LinearMixtureModel

getDistributionCount, getDistributions, getPriorWeights, getPriorWeightSum, sample, sampleInto, setDistributions, setPriorWeights, toString

Methods inherited from class gov.sandia.cognition.statistics.AbstractDistribution

sample

Methods inherited from class java.lang.Object

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

Methods inherited from interface gov.sandia.cognition.statistics.Distribution

sample, sample, sampleInto

Constructor Detail

ScalarMixtureDensityModel

public ScalarMixtureDensityModel()

Creates a new instance of ScalarMixtureDensityModel

ScalarMixtureDensityModel

public ScalarMixtureDensityModel(SmoothUnivariateDistribution... distributions)

Creates a new instance of ScalarMixtureDensityModel

Parameters:

distributions - Distributions that comprise the SMDM with equal prior weight

ScalarMixtureDensityModel

public ScalarMixtureDensityModel(java.util.Collection<? extends SmoothUnivariateDistribution> distributions)

Creates a new instance of ScalarMixtureDensityModel

Parameters:

distributions - Distributions that comprise the SMDM with equal prior weight

ScalarMixtureDensityModel

public ScalarMixtureDensityModel(java.util.Collection<? extends SmoothUnivariateDistribution> distributions,
                                 double[] priorWeights)

Creates a new instance of ScalarMixtureDensityModel

Parameters:

distributions - Distributions that comprise the SMDM

priorWeights - Weights proportionate by which the distributions are sampled

ScalarMixtureDensityModel

public ScalarMixtureDensityModel(ScalarMixtureDensityModel other)

Copy constructor

Parameters:

other - SMDM to copy

Method Detail

clone

public ScalarMixtureDensityModel clone()

Description copied from class: AbstractCloneableSerializable

This makes public the clone method on the Object class and removes the exception that it throws. Its default behavior is to automatically create a clone of the exact type of object that the clone is called on and to copy all primitives but to keep all references, which means it is a shallow copy. Extensions of this class may want to override this method (but call super.clone() to implement a "smart copy". That is, to target the most common use case for creating a copy of the object. Because of the default behavior being a shallow copy, extending classes only need to handle fields that need to have a deeper copy (or those that need to be reset). Some of the methods in ObjectUtil may be helpful in implementing a custom clone method. Note: The contract of this method is that you must use super.clone() as the basis for your implementation.

Specified by:

clone in interface Vectorizable

Specified by:

clone in interface CloneableSerializable

Overrides:

clone in class LinearMixtureModel<java.lang.Double,SmoothUnivariateDistribution>

Returns:

A clone of this object.

convertToVector

public Vector convertToVector()

Description copied from interface: Vectorizable

Converts the object to a vector.

Specified by:

convertToVector in interface Vectorizable

Returns:

The Vector form of the object.

convertFromVector

public void convertFromVector(Vector parameters)

Description copied from interface: Vectorizable

Converts the object from a Vector of parameters.

Specified by:

convertFromVector in interface Vectorizable

Parameters:

parameters - The parameters to incorporate.

getMinSupport

public java.lang.Double getMinSupport()

Description copied from interface: UnivariateDistribution

Gets the minimum support (domain or input) of the distribution.

Specified by:

getMinSupport in interface UnivariateDistribution<java.lang.Double>

Returns:

Minimum support.

getMaxSupport

public java.lang.Double getMaxSupport()

Description copied from interface: UnivariateDistribution

Gets the minimum support (domain or input) of the distribution.

Specified by:

getMaxSupport in interface UnivariateDistribution<java.lang.Double>

Returns:

Minimum support.

getMean

public java.lang.Double getMean()

Description copied from interface: DistributionWithMean

Gets the arithmetic mean, or "first central moment" or "expectation", of the distribution.

Specified by:

getMean in interface DistributionWithMean<java.lang.Double>

Specified by:

getMean in interface SmoothUnivariateDistribution

Returns:

Mean of the distribution.

getMeanAsDouble

public double getMeanAsDouble()

Description copied from interface: UnivariateDistribution

Gets the mean of the distribution as a double.

Specified by:

getMeanAsDouble in interface UnivariateDistribution<java.lang.Double>

Returns:

The mean as a double.

getVariance

@PublicationReference(author="Wikipedia",
                      title="Mixture Model",
                      type=WebPage,
                      year=2009,
                      url="http://en.wikipedia.org/wiki/Mixture_density")
public double getVariance()

Description copied from interface: UnivariateDistribution

Gets the variance of the distribution. This is sometimes called the second central moment by more pedantic people, which is equivalent to the square of the standard deviation.

Specified by:

getVariance in interface UnivariateDistribution<java.lang.Double>

Returns:

Variance of the distribution.

sampleAsDouble

public double sampleAsDouble(java.util.Random random)

Description copied from interface: SmoothUnivariateDistribution

Samples a value from this distribution as a double. This is a convenience method to potentially avoid boxing.

Specified by:

sampleAsDouble in interface SmoothUnivariateDistribution

Parameters:

random - Random number generator to use.

Returns:

A value sampled from this distribution.

sampleAsDoubles

public double[] sampleAsDoubles(java.util.Random random,
                                int count)

Description copied from interface: SmoothUnivariateDistribution

Samples values from this distribution as an array of doubles. This is a convenience method to potentially avoid boxing.

Specified by:

sampleAsDoubles in interface SmoothUnivariateDistribution

Parameters:

random - Random number generator to use.

count - The number of values to sample. Cannot be negative

Returns:

An array of values sampled from this distribution. Size is that of count.

sampleInto

public void sampleInto(java.util.Random random,
                       double[] output,
                       int start,
                       int length)

Description copied from interface: SmoothUnivariateDistribution

Samples values from this distribution as an array of doubles. This is a convenience method to potentially avoid boxing.

Specified by:

sampleInto in interface SmoothUnivariateDistribution

Parameters:

random - Random number generator to use.

output - The array to write the result into. Cannot be null.

start - The offset in the array to start writing at. Cannot be negative.

length - The number of values to sample. Cannot be negative.

getProbabilityFunction

public ScalarMixtureDensityModel.PDF getProbabilityFunction()

Description copied from interface: ComputableDistribution

Gets the distribution function associated with this Distribution, either the PDF or PMF.

Specified by:

getProbabilityFunction in interface ComputableDistribution<java.lang.Double>

Specified by:

getProbabilityFunction in interface SmoothUnivariateDistribution

Returns:

Distribution function associated with this Distribution.

getCDF

public ScalarMixtureDensityModel.CDF getCDF()

Description copied from interface: UnivariateDistribution

Gets the CDF of a scalar distribution.

Specified by:

getCDF in interface ClosedFormUnivariateDistribution<java.lang.Double>

Specified by:

getCDF in interface SmoothUnivariateDistribution

Specified by:

getCDF in interface UnivariateDistribution<java.lang.Double>

Returns:

CDF of the scalar distribution.