edu.ucla.stat.SOCR.core
Class Modeler

java.lang.Object
  edu.ucla.stat.SOCR.core.Modeler

public class Modeler
extends java.lang.Object

Field Summary

static int	CONTINUOUS_DISTRIBUTION_TYPE
static int	DISCRETE_DISTRIBUTION_TYPE
static int	FOURIER_TYPE
static int	MIXED_DISTRIBUTION_TYPE
static int	WAVELET_TYPE

Constructor Summary

Modeler()

Method Summary

void	fitCurve(float[] rawDat, double minx, double maxx, javax.swing.JTextArea resultPanelTextArea, boolean rescaleClicked, boolean scaleUp, boolean initReset) takes data along with x, y limits and fits a pdf to the data range and stores the resulting model fit in data arrays that must be returned by calls to returnModelX() and returnModelY()
double[]	generateSamples(int sampleCount) generates samples from the distribution and returns a double[] data type
java.lang.String	getDescription() return the description for this modeler
double	getGraphLowerLimit()
double	getGraphUpperLimit()
java.lang.String	getInstructions() return the instructions for using this modeler
java.lang.String	getKSModelTestString(java.lang.String distributionModelName, int numberOfQuantiles, double[] x, double[] y) getKSModelTestString computes the KolmogorovSmirnoff test statistics of the match between the 100 quantiles of the data and their corresponding model-distribution quartile counterparts!
double	getLowerLimit() return the allowable x limit values.
int	getModelCount() return the number of models to be plotted.
int	getModelType() returns one if model is of distribution and needs to be rescaled for display.
java.lang.String	getResearch() return the references for this modeler
double	getUpperLimit() return the allowaable x limit values.
boolean	isContinuous()
void	registerObservers(ObservableWrapper o)
double[]	returnModelX() returns the fitted model values for X axis
double[]	returnModelY() returns the fitted densisty for corresponding X axis values
boolean	useInitButton() What is this method used for?

Methods inherited from class java.lang.Object

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

Field Detail

CONTINUOUS_DISTRIBUTION_TYPE

public static final int CONTINUOUS_DISTRIBUTION_TYPE

See Also:

Constant Field Values

DISCRETE_DISTRIBUTION_TYPE

public static final int DISCRETE_DISTRIBUTION_TYPE

See Also:

Constant Field Values

MIXED_DISTRIBUTION_TYPE

public static final int MIXED_DISTRIBUTION_TYPE

See Also:

Constant Field Values

FOURIER_TYPE

public static final int FOURIER_TYPE

See Also:

Constant Field Values

WAVELET_TYPE

public static final int WAVELET_TYPE

See Also:

Constant Field Values

Constructor Detail

Modeler

public Modeler()

Method Detail

fitCurve

public void fitCurve(float[] rawDat,
                     double minx,
                     double maxx,
                     javax.swing.JTextArea resultPanelTextArea,
                     boolean rescaleClicked,
                     boolean scaleUp,
                     boolean initReset)

takes data along with x, y limits and fits a pdf to the data range and stores the resulting model fit in data arrays that must be returned by calls to returnModelX() and returnModelY()

generateSamples

public double[] generateSamples(int sampleCount)

generates samples from the distribution and returns a double[] data type

getDescription

public java.lang.String getDescription()

return the description for this modeler

getGraphUpperLimit

public double getGraphUpperLimit()

getGraphLowerLimit

public double getGraphLowerLimit()

getInstructions

public java.lang.String getInstructions()

return the instructions for using this modeler

getKSModelTestString

public java.lang.String getKSModelTestString(java.lang.String distributionModelName,
                                             int numberOfQuantiles,
                                             double[] x,
                                             double[] y)

getKSModelTestString computes the KolmogorovSmirnoff test statistics of the match between the 100 quantiles of the data and their corresponding model-distribution quartile counterparts!

Parameters:

distributionModelName - name of the distribution model-fit

numberOfQuantiles - number of quantiles to use (e.g., 100)

x - data quantiles array

y - model quantiles array

getLowerLimit

public double getLowerLimit()

return the allowable x limit values. This method should return the lower limit. eg: for a normal distribution lowerlimit = NEGATIVE_INFINITY

getModelCount

public int getModelCount()

return the number of models to be plotted. If n models are returned, the vectors from returnModelX and returnModelY will be split into n equal sub sections and plotted

Returns:

getModelType

public int getModelType()

returns one if model is of distribution and needs to be rescaled for display. Returns 0 if model does not require scaling. EG: Polynomial fit is type 0 and normal distribution is type 1.

getResearch

public java.lang.String getResearch()

return the references for this modeler

getUpperLimit

public double getUpperLimit()

return the allowaable x limit values. This method should return the upper limit. eg: for a normal distribution upperlimit = POSITIVE_INFINITY

isContinuous

public boolean isContinuous()

registerObservers

public void registerObservers(ObservableWrapper o)

returnModelX

public double[] returnModelX()

returns the fitted model values for X axis

returnModelY

public double[] returnModelY()

returns the fitted densisty for corresponding X axis values

useInitButton

public boolean useInitButton()

What is this method used for?