repicea.simulation

Class REpiceaPredictor

java.lang.Object

repicea.simulation.SensitivityAnalysisParameter<ModelParameterEstimates>

repicea.simulation.REpiceaPredictor

All Implemented Interfaces:

java.io.Serializable

Direct Known Subclasses:

AbstractStemTaperPredictor, DopalepDbhIncrementPredictor, FrenchCommercialVolume2014Predictor, FrenchCommercialVolume2020Predictor, FrenchHDRelationship2014Predictor, FrenchHDRelationship2018Predictor, HDRelationshipPredictor, MathildeClimatePredictor, MathildeClimatePredictor, MathildeDiameterIncrementPredictor, MathildeRecruitDbhPredictor, MathildeRecruitmentNumberPredictor, MathildeSubModule, REpiceaBinaryEventPredictor

public abstract class REpiceaPredictor
extends SensitivityAnalysisParameter<ModelParameterEstimates>

This class is the basic class for all models that are designed for predictions. It implements the methods for stochastic simulations.

See Also:

Serialized Form

Nested Class Summary

Nested Classes

Modifier and Type	Class and Description
protected static class	REpiceaPredictor.CruiseLine This class creates a fake subject for cruise line random effects.
static class	REpiceaPredictor.ErrorTermGroup
protected static class	REpiceaPredictor.IntervalNestedInPlotDefinition This class creates a fake subject for interval random effects nested in the plots.

Field Summary

Fields

Modifier and Type	Field and Description
protected static java.util.List<java.lang.Integer>	DefaultZeroIndex
protected boolean	isRandomEffectsVariabilityEnabled
protected boolean	isResidualVariabilityEnabled
protected java.util.concurrent.CopyOnWriteArrayList<REpiceaPredictorListener>	listeners
protected Matrix	oXVector

Fields inherited from class repicea.simulation.SensitivityAnalysisParameter

isParametersVariabilityEnabled

Constructor Summary

Constructors

Modifier	Constructor and Description
protected	REpiceaPredictor(boolean isParametersVariabilityEnabled, boolean isRandomEffectsVariabilityEnabled, boolean isResidualVariabilityEnabled) General constructor for all combinations of uncertainty sources.

Method Summary

Modifier and Type	Method and Description
void	addModelBasedSimulatorListener(REpiceaPredictorListener listener) This method adds the listener instance to the list of listeners.
protected boolean	doBlupsExistForThisSubject(MonteCarloSimulationCompliantObject subject)
protected boolean	doesThisSubjectHaveResidualErrorTerm(MonteCarloSimulationCompliantObject subject)
protected boolean	doRandomDeviatesExistForThisSubject(MonteCarloSimulationCompliantObject subject)
protected void	fireModelBasedSimulatorEvent(REpiceaPredictorEvent event)
protected void	fireRandomEffectDeviateGeneratedEvent(MonteCarloSimulationCompliantObject subject, Estimate<? extends StandardGaussianDistribution> originalRandomEffects, Matrix randomDeviates)
protected Estimate<? extends StandardGaussianDistribution>	getBlupsForThisSubject(MonteCarloSimulationCompliantObject subject) This method returns the blups for the subject or nothing if there is no blups for this subject
protected REpiceaPredictor.CruiseLine	getCruiseLineForThisSubject(java.lang.String cruiseLineID, MonteCarloSimulationCompliantObject stand) This method checks if a cruise line exists for this plot
protected java.util.Map<java.lang.String,Estimate<? extends StandardGaussianDistribution>>	getDefaultRandomEffects()
protected Estimate<? extends StandardGaussianDistribution>	getDefaultRandomEffects(HierarchicalLevel level)
protected GaussianErrorTermEstimate	getDefaultResidualError(java.lang.Enum<?> enumVar)
protected GaussianErrorTermList	getGaussianErrorTerms(MonteCarloSimulationCompliantObject subject)
protected REpiceaPredictor.IntervalNestedInPlotDefinition	getIntervalNestedInPlotDefinition(MonteCarloSimulationCompliantObject stand, int date) This method checks if the interval definition is available for the stand at that date.
protected Matrix	getParametersForThisRealization(MonteCarloSimulationCompliantObject subject) This method calls the setSpecificParametersDeviateForThisRealization method if the parameter variability is enabled and returns a realization-specific simulated vector of model parameters.
protected Matrix	getRandomEffectsForThisSubject(MonteCarloSimulationCompliantObject subject) This method calls the setSpecificPlotRandomEffectsForThisStand method if the random effects variability is enabled and returns a stand-specific simulated vector of random effects.
protected Matrix	getResidualError() This method returns the residual error under the assumption of iid and that the error is unique for all groups.
protected Matrix	getResidualErrorForThisSubject(MonteCarloSimulationCompliantObject subject, java.lang.Enum<?> group) This method returns the residual error or the vector of residual errors associated with the subjectId.
protected java.lang.String	getSubjectPlusMonteCarloSpecificId(MonteCarloSimulationCompliantObject object)
protected static java.lang.String	getSubjectPlusMonteCarloSpecificId(java.lang.String subjectID, int monteCarloRealizationID)
protected boolean	hasSubjectBeenTestedForBlups(MonteCarloSimulationCompliantObject subject)
protected abstract void	init() This method reads all the parameters in .csv files and stores the estimates into members defaultBeta, defaultResidualError, and defaultRandomEffects.
protected void	recordSubjectTestedForBlups(MonteCarloSimulationCompliantObject subject)
void	removeModelBasedSimulatorListener(REpiceaPredictorListener listener) This method removes the listener instance from the list of listeners.
protected void	setBlupsForThisSubject(MonteCarloSimulationCompliantObject subject, Estimate<? extends StandardGaussianDistribution> blups)
protected void	setDefaultRandomEffects(HierarchicalLevel level, Estimate<? extends StandardGaussianDistribution> newEstimate)
protected void	setDefaultResidualError(java.lang.Enum<?> enumVar, GaussianErrorTermEstimate estimate)
protected void	setDeviatesForRandomEffectsOfThisSubject(MonteCarloSimulationCompliantObject subject, Matrix randomDeviates)
protected void	setParameterEstimates(ModelParameterEstimates gaussianEstimate)
protected Matrix	simulateDeviatesForRandomEffectsOfThisSubject(MonteCarloSimulationCompliantObject subject, Estimate<?> randomEffectsEstimate) This method simulates random deviates from an estimate and stores them in the simulatedRandomEffects member.

Methods inherited from class repicea.simulation.SensitivityAnalysisParameter

getParameterEstimates

Methods inherited from class java.lang.Object

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

Field Detail

DefaultZeroIndex

protected static final java.util.List<java.lang.Integer> DefaultZeroIndex

listeners

protected final java.util.concurrent.CopyOnWriteArrayList<REpiceaPredictorListener> listeners

isRandomEffectsVariabilityEnabled

protected boolean isRandomEffectsVariabilityEnabled

isResidualVariabilityEnabled

protected boolean isResidualVariabilityEnabled

oXVector

protected Matrix oXVector

Constructor Detail

REpiceaPredictor

protected REpiceaPredictor(boolean isParametersVariabilityEnabled,
                           boolean isRandomEffectsVariabilityEnabled,
                           boolean isResidualVariabilityEnabled)

General constructor for all combinations of uncertainty sources.

Parameters:

isParametersVariabilityEnabled - a boolean that enables the variability at the parameter level

isRandomEffectsVariabilityEnabled - a boolean that enables the variability at the random effect level

isResidualVariabilityEnabled - a boolean that enables the variability at the tree level

Method Detail

init

protected abstract void init()

This method reads all the parameters in .csv files and stores the estimates into members defaultBeta, defaultResidualError, and defaultRandomEffects.

getDefaultRandomEffects

protected java.util.Map<java.lang.String,Estimate<? extends StandardGaussianDistribution>> getDefaultRandomEffects()

setParameterEstimates

protected void setParameterEstimates(ModelParameterEstimates gaussianEstimate)

Overrides:

setParameterEstimates in class SensitivityAnalysisParameter<ModelParameterEstimates>

setDefaultRandomEffects

protected void setDefaultRandomEffects(HierarchicalLevel level,
                                       Estimate<? extends StandardGaussianDistribution> newEstimate)

getDefaultRandomEffects

protected Estimate<? extends StandardGaussianDistribution> getDefaultRandomEffects(HierarchicalLevel level)

setDefaultResidualError

protected void setDefaultResidualError(java.lang.Enum<?> enumVar,
                                       GaussianErrorTermEstimate estimate)

getDefaultResidualError

protected GaussianErrorTermEstimate getDefaultResidualError(java.lang.Enum<?> enumVar)

getIntervalNestedInPlotDefinition

protected REpiceaPredictor.IntervalNestedInPlotDefinition getIntervalNestedInPlotDefinition(MonteCarloSimulationCompliantObject stand,
                                                                                            int date)

This method checks if the interval definition is available for the stand at that date. If it is, it returns the instance. Otherwise, it creates a new interval definition.

Parameters:

stand - A MonteCarloSimulationCompliantObject that designates the stand

date - an Integer

Returns:

an IntervalDefinition instance

getCruiseLineForThisSubject

protected REpiceaPredictor.CruiseLine getCruiseLineForThisSubject(java.lang.String cruiseLineID,
                                                                  MonteCarloSimulationCompliantObject stand)

This method checks if a cruise line exists for this plot

Parameters:

cruiseLineID - the id of the cruise line

stand - a MonteCarloSimulationCompliantObject instance

Returns:

a CruiseLine instance

getParametersForThisRealization

protected final Matrix getParametersForThisRealization(MonteCarloSimulationCompliantObject subject)

This method calls the setSpecificParametersDeviateForThisRealization method if the parameter variability is enabled and returns a realization-specific simulated vector of model parameters. Otherwise it returns a default vector (beta). Note that the simulated parameters are related to the Monte Carlo realization. For instance, all subject in a given Monte Carlo realization will have the same simulation parameters.

Overrides:

getParametersForThisRealization in class SensitivityAnalysisParameter<ModelParameterEstimates>

Parameters:

subject - a subject that implements the MonteCarloSimulationCompliantObject interface

Returns:

a vector of parameters

fireRandomEffectDeviateGeneratedEvent

protected void fireRandomEffectDeviateGeneratedEvent(MonteCarloSimulationCompliantObject subject,
                                                     Estimate<? extends StandardGaussianDistribution> originalRandomEffects,
                                                     Matrix randomDeviates)

simulateDeviatesForRandomEffectsOfThisSubject

protected Matrix simulateDeviatesForRandomEffectsOfThisSubject(MonteCarloSimulationCompliantObject subject,
                                                               Estimate<?> randomEffectsEstimate)

This method simulates random deviates from an estimate and stores them in the simulatedRandomEffects member.

Parameters:

subject - a MonteCarloSimulationCompliantObject instance

randomEffectsEstimate - the estimate from which the random deviates are generated

Returns:

the random deviates as a Matrix instance (a copy of it)

setDeviatesForRandomEffectsOfThisSubject

protected final void setDeviatesForRandomEffectsOfThisSubject(MonteCarloSimulationCompliantObject subject,
                                                              Matrix randomDeviates)

getSubjectPlusMonteCarloSpecificId

protected final java.lang.String getSubjectPlusMonteCarloSpecificId(MonteCarloSimulationCompliantObject object)

getSubjectPlusMonteCarloSpecificId

protected static java.lang.String getSubjectPlusMonteCarloSpecificId(java.lang.String subjectID,
                                                                     int monteCarloRealizationID)

getRandomEffectsForThisSubject

protected final Matrix getRandomEffectsForThisSubject(MonteCarloSimulationCompliantObject subject)

This method calls the setSpecificPlotRandomEffectsForThisStand method if the random effects variability is enabled and returns a stand-specific simulated vector of random effects. Otherwise it returns a default vector (all elements set to 0).

Parameters:

subject - a MonteCarloSimulationCompliantObject object

Returns:

a Matrix object

doRandomDeviatesExistForThisSubject

protected final boolean doRandomDeviatesExistForThisSubject(MonteCarloSimulationCompliantObject subject)

getResidualErrorForThisSubject

protected final Matrix getResidualErrorForThisSubject(MonteCarloSimulationCompliantObject subject,
                                                      java.lang.Enum<?> group)

This method returns the residual error or the vector of residual errors associated with the subjectId. If the subject parameter is entered as null, the method assumes there is no need to store the simulated error terms in the simulatedResidualError map. This feature is useful if the residual error terms are identically and independently distributed.

Parameters:

subject - a MonteCarloSimulationCompliantObject instance

group - an Enum that defines the group in case of different error term specifications

Returns:

a Matrix instance

getGaussianErrorTerms

protected final GaussianErrorTermList getGaussianErrorTerms(MonteCarloSimulationCompliantObject subject)

doesThisSubjectHaveResidualErrorTerm

protected final boolean doesThisSubjectHaveResidualErrorTerm(MonteCarloSimulationCompliantObject subject)

getResidualError

protected final Matrix getResidualError()

This method returns the residual error under the assumption of iid and that the error is unique for all groups. See the getResidualErrorForThisSubject method.

Returns:

a Matrix instance

fireModelBasedSimulatorEvent

protected void fireModelBasedSimulatorEvent(REpiceaPredictorEvent event)

addModelBasedSimulatorListener

public void addModelBasedSimulatorListener(REpiceaPredictorListener listener)

This method adds the listener instance to the list of listeners.

Parameters:

listener - a ModelBasedSimulatorListener listener

removeModelBasedSimulatorListener

public void removeModelBasedSimulatorListener(REpiceaPredictorListener listener)

This method removes the listener instance from the list of listeners.

Parameters:

listener - a ModelBasedSimulatorListener listener

doBlupsExistForThisSubject

protected boolean doBlupsExistForThisSubject(MonteCarloSimulationCompliantObject subject)

getBlupsForThisSubject

protected Estimate<? extends StandardGaussianDistribution> getBlupsForThisSubject(MonteCarloSimulationCompliantObject subject)

This method returns the blups for the subject or nothing if there is no blups for this subject

Parameters:

subject - a MonteCarloSimulationCompliantObject instance

Returns:

an Estimate instance or null

setBlupsForThisSubject

protected final void setBlupsForThisSubject(MonteCarloSimulationCompliantObject subject,
                                            Estimate<? extends StandardGaussianDistribution> blups)

recordSubjectTestedForBlups

protected final void recordSubjectTestedForBlups(MonteCarloSimulationCompliantObject subject)

hasSubjectBeenTestedForBlups

protected final boolean hasSubjectBeenTestedForBlups(MonteCarloSimulationCompliantObject subject)