desmoj.core.statistic

Class Regression

java.lang.Object

desmoj.core.simulator.NamedObject

desmoj.core.simulator.ModelComponent

desmoj.core.simulator.Reportable

desmoj.core.statistic.StatisticObject

desmoj.core.statistic.Regression

All Implemented Interfaces:

java.util.Observer

public class Regression
extends StatisticObject

The Regression class is producing a linear regression for two ValueSupplier objects called x and y.

Version:

DESMO-J, Ver. 2.5.1c copyright (c) 2015

Author:

Soenke Claassen, based on DESMO-C from Thomas Schniewind, 1998 Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with the License. You may obtain a copy of the License at http://www.apache.org/licenses/LICENSE-2.0 Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the specific language governing permissions and limitations under the License.

Field Summary

Fields inherited from class desmoj.core.statistic.StatisticObject

UNDEFINED

Constructor Summary

Constructors

Constructor and Description
Regression(Model ownerModel, java.lang.String name, java.lang.String xName, java.lang.String yName, boolean showInReport, boolean showInTrace) Constructor for a Regression object that is NOT connected to any ValueSuppliers for the x and y values.
Regression(Model ownerModel, java.lang.String name, java.lang.String xName, java.lang.String yName, ValueSupplier xValsup, ValueSupplier yValsup, boolean showInReport, boolean showInTrace) Constructor for a Regression object that is observing two ValueSuppliers for the x and y values.
Regression(Model ownerModel, java.lang.String name, ValueSupplier xValsup, ValueSupplier yValsup, boolean showInReport, boolean showInTrace) Constructor for a Regression object with no names for the x and y values.

Method Summary

Modifier and Type	Method and Description
double	correlationCoeff() Returns the correlation coefficient.
Reporter	createDefaultReporter() Returns a Reporter to produce a report about this Regression analysis.
double	getXMean() Returns the mean x-value.
java.lang.String	getXName() Returns the name of the x-value.
double	getXValue() Returns the actual value of the ValueSupplier supplying the x-value.
double	getYMean() Returns the mean y-value.
java.lang.String	getYName() Returns the name of the y-value.
double	getYValue() Returns the actual value of the ValueSupplier supplying the y-value.
double	intercept() Returns the interception of the X-axis.
double	regCoeff() Returns the regression coefficient.
void	reset() Resets this Regression object by resetting all variables of x and y to 0.0 .
double	residualStdDev() Returns the residuale standard deviation.
double	stdDevRegCoeff() Returns the standard deviation of the regression coefficients.
void	update() Updates this Regression object by fetching the actual values of the ValueSupplier s and processing them.
void	update(double xVal, double yVal) Updates this Regression object with the x and y double values given as parameters.
void	update(java.util.Observable o, java.lang.Object arg) Implementation of the virtual update(Observable, Object) method of the Observer interface.
boolean	xIsConstant() Returns true when the x-values are constant.
boolean	yIsConstant() Returns true when the y-values are constant.

Methods inherited from class desmoj.core.statistic.StatisticObject

getUnit, getUnitText, round, setUnit

Methods inherited from class desmoj.core.simulator.Reportable

getCorrespondingSchedulable, getDescription, getObservations, getReporter, incrementObservations, incrementObservations, reportIsOn, reportOff, reportOn, resetAt, setCorrespondingSchedulable, setDescription, setReporter

Methods inherited from class desmoj.core.simulator.ModelComponent

current, currentEntity, currentEntityAll, currentEvent, currentModel, currentSimProcess, debugIsOn, debugOff, debugOn, getModel, isExperimentCompatible, isModelCompatible, presentTime, sendDebugNote, sendMessage, sendTraceNote, sendWarning, skipTraceNote, skipTraceNote, traceIsOn, traceOff, traceOn

Methods inherited from class desmoj.core.simulator.NamedObject

getName, getQuotedName, toString

Methods inherited from class java.lang.Object

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

Constructor Detail

Regression

public Regression(Model ownerModel,
                  java.lang.String name,
                  ValueSupplier xValsup,
                  ValueSupplier yValsup,
                  boolean showInReport,
                  boolean showInTrace)

Constructor for a Regression object with no names for the x and y values. The names will be set to "X" and "Y".

Regression

public Regression(Model ownerModel,
                  java.lang.String name,
                  java.lang.String xName,
                  java.lang.String yName,
                  ValueSupplier xValsup,
                  ValueSupplier yValsup,
                  boolean showInReport,
                  boolean showInTrace)

Constructor for a Regression object that is observing two ValueSuppliers for the x and y values.

Parameters:

ownerModel - Model : The model this Regression object is associated to

name - java.lang.String : The name of this Regression object

xName - java.lang.String : The name for the x values.

yName - java.lang.String : The name for the y values.

xValsup - ValueSupplier : The ValueSupplier providing the x values for the regression analysis. The given ValueSupplier will be observed by this Regression object.

yValsup - ValueSupplier : The ValueSupplier providing the y values for the regression analysis. The given ValueSupplier will be observed by this Regression object.

showInReport - boolean : Flag for showing the report about this Regression.

showInTrace - boolean : Flag for showing the trace output of this Regression.

Regression

public Regression(Model ownerModel,
                  java.lang.String name,
                  java.lang.String xName,
                  java.lang.String yName,
                  boolean showInReport,
                  boolean showInTrace)

Constructor for a Regression object that is NOT connected to any ValueSuppliers for the x and y values.

Parameters:

ownerModel - Model : The model this Regression object is associated to

name - java.lang.String : The name of this Regression object

xName - java.lang.String : The name for the x values.

yName - java.lang.String : The name for the y values.

showInReport - boolean : Flag for showing the report about this Regression.

showInTrace - boolean : Flag for showing the trace output of this Regression.

Method Detail

correlationCoeff

public double correlationCoeff()

Returns the correlation coefficient.

Returns:

double : The correlation coefficient.

createDefaultReporter

public Reporter createDefaultReporter()

Returns a Reporter to produce a report about this Regression analysis.

Returns:

desmoj.report.Reporter : The Reporter for this Regression.

getXMean

public double getXMean()

Returns the mean x-value.

Returns:

double : The mean x-value.

getXName

public java.lang.String getXName()

Returns the name of the x-value.

Returns:

String : The name of the x-value.

getXValue

public double getXValue()

Returns the actual value of the ValueSupplier supplying the x-value.

Returns:

double : The actual value of the observed ValueSupplier object supplying the x-value.

getYMean

public double getYMean()

Returns the mean y-value.

Returns:

double : The mean y-value.

getYName

public java.lang.String getYName()

Returns the name of the y-value.

Returns:

String : The name of the y-value.

getYValue

public double getYValue()

Returns the actual value of the ValueSupplier supplying the y-value.

Returns:

double : The actual value of the observed ValueSupplier object supplying the y-value.

intercept

public double intercept()

Returns the interception of the X-axis.

Returns:

double : The interception of the X-axis.

regCoeff

public double regCoeff()

Returns the regression coefficient.

Returns:

double : The regression coefficient.

reset

public void reset()

Resets this Regression object by resetting all variables of x and y to 0.0 . The names remain the same.

Overrides:

reset in class Reportable

residualStdDev

public double residualStdDev()

Returns the residuale standard deviation.

Returns:

double : The residuale standard deviation.

stdDevRegCoeff

public double stdDevRegCoeff()

Returns the standard deviation of the regression coefficients.

Returns:

double : The standard deviation of the regression coefficients.

update

public void update()

Updates this Regression object by fetching the actual values of the ValueSupplier s and processing them. The ValueSupplier s are passed in the constructor of this Regression object. This update() method complies with the one described in DESMO, see [Page91].

update

public void update(double xVal,
                   double yVal)

Updates this Regression object with the x and y double values given as parameters. In some cases it might be more convenient to pass the values this Regression will be updated with directly within the update(double xVal, double yVal) method instead of going via the ValueSupplier.

Parameters:

xVal - double : The x value with which this Regression will be updated.

yVal - double : The y value with which this Regression will be updated.

update

public void update(java.util.Observable o,
                   java.lang.Object arg)

Implementation of the virtual update(Observable, Object) method of the Observer interface. This method will be called automatically from an Observable object within its notifyObservers() method.
Whenever either of the two ValueSupplier objects providing the x or y value is updated and calling notifyStatistics() the x and the y value will be fetched and a new regression will be calculated.

Parameters:

o - java.util.Observable : The Observable calling this method within its own notifyObservers() method.

arg - Object : The Object with which this Regression is updated.

xIsConstant

public boolean xIsConstant()

Returns true when the x-values are constant. That means no considerable change can be recognized.

Returns:

boolean :true is returned if no considerable change of the x-values can be recognized. false otherwise.

yIsConstant

public boolean yIsConstant()

Returns true when the y-values are constant. That means no considerable change can be recognized.

Returns:

boolean :true is returned if no considerable change of the y-values can be recognized. false otherwise.