Package pulse.search.statistics

Class ModelSelectionCriterion

java.lang.Object

pulse.util.UpwardsNavigable

pulse.util.Group

pulse.util.Accessible

pulse.util.PropertyHolder

pulse.search.statistics.Statistic

pulse.search.statistics.ModelSelectionCriterion

All Implemented Interfaces:

Serializable, Descriptive, Reflexive

Direct Known Subclasses:

AICStatistic, BICStatistic

public abstract class ModelSelectionCriterion
extends Statistic

An abstract superclass for the BIC and AIC statistics.

See Also:

Serialized Form

Constructor Summary

Constructors

Constructor	Description
ModelSelectionCriterion(ModelSelectionCriterion another)
ModelSelectionCriterion(OptimiserStatistic os)

Method Summary

Modifier and Type	Method	Description
void	calcCriterion()	This calculates either the AIC or BIC statistic, which only differ by the penalising term.
abstract ModelSelectionCriterion	copy()
boolean	equals(Object obj)
void	evaluate(GeneralTask t)
int	getNumVariables()
OptimiserStatistic	getOptimiserStatistic()
NumericProperty	getStatistic()
int	hashCode()
abstract double	penalisingTerm(int k, int n)	The penalising term, which is different depending on implementation.
double	probability(List<ModelSelectionCriterion> all)	Calculates the probability that this model is the best among all others.
void	set(NumericPropertyKeyword key, NumericProperty p)	An abstract method, which must be overriden to gain access over setting the values of all relevant (selected by the programmer) NumericProperties in subclasses of Accessible.
void	setOptimiserStatistic(OptimiserStatistic os)
void	setStatistic(NumericProperty p)
NumericProperty	weight(List<ModelSelectionCriterion> all)	Calculates the weight (in the Akaike sense) when comparing the model associated with this statistic with other models represented by statistics of the same type.

Methods inherited from class pulse.util.PropertyHolder

addListener, areDetailsHidden, data, describe, firePropertyChanged, getDescriptor, getListeners, getPrefix, ignoreSiblings, initListeners, isListedNumericType, isListedParameter, listedKeywords, listedTypes, numericData, parameterListChanged, removeListeners, setPrefix, updateProperties, updateProperty

Methods inherited from class pulse.util.Accessible

accessibleChildren, genericProperties, genericProperty, numericProperties, numericProperty, property, update, update

Methods inherited from class pulse.util.Group

access, children, contents, getSimpleName, subgroups

Methods inherited from class pulse.util.UpwardsNavigable

addHierarchyListener, getHierarchyListeners, getParent, identify, removeHierarchyListener, removeHierarchyListeners, setParent, specificAncestor, tellParent

Methods inherited from class java.lang.Object

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

Constructor Detail

ModelSelectionCriterion

public ModelSelectionCriterion(OptimiserStatistic os)

ModelSelectionCriterion

public ModelSelectionCriterion(ModelSelectionCriterion another)

Method Detail

hashCode

public int hashCode()

Overrides:

hashCode in class Object

equals

public boolean equals(Object obj)

Overrides:

equals in class Object

evaluate

public void evaluate(GeneralTask t)

Specified by:

evaluate in class Statistic

calcCriterion

public void calcCriterion()

This calculates either the AIC or BIC statistic, which only differ by the penalising term.

See Also:

penalisingTerm()

penalisingTerm

public abstract double penalisingTerm(int k,
                                      int n)

The penalising term, which is different depending on implementation.

Parameters:

k - the number of model variables

n - the sample size

Returns:

the penalising term

copy

public abstract ModelSelectionCriterion copy()

weight

public NumericProperty weight(List<ModelSelectionCriterion> all)

Calculates the weight (in the Akaike sense) when comparing the model associated with this statistic with other models represented by statistics of the same type.

Parameters:

the - selection statistics of the same type as this one

Returns:

a NumericProperty of the MODEL_WEIGHT type, which is the probability this model is the best one.

probability

public double probability(List<ModelSelectionCriterion> all)

Calculates the probability that this model is the best among all others.

Parameters:

all - statistics from models that will be compared with this one

Returns:

the probability, which is a decimal value within the [0,1] range.

getNumVariables

public int getNumVariables()

getOptimiserStatistic

public OptimiserStatistic getOptimiserStatistic()

setOptimiserStatistic

public final void setOptimiserStatistic(OptimiserStatistic os)

setStatistic

public final void setStatistic(NumericProperty p)

getStatistic

public NumericProperty getStatistic()

set

public void set(NumericPropertyKeyword key,
                NumericProperty p)

Description copied from class: Accessible

An abstract method, which must be overriden to gain access over setting the values of all relevant (selected by the programmer) NumericProperties in subclasses of Accessible. Typically this involves a switch statement that goes through the different options for the type and invokes different set(...) methods to update the matching NumericProperty with property.

Specified by:

set in class Accessible

Parameters:

key - the type, which must be equal by definition to property.getType().

p - the property, which contains new information.