rvl.datasim
Class Factor

java.lang.Object
  |
  +--rvl.datasim.Predictor
        |
        +--rvl.datasim.Factor

All Implemented Interfaces:

java.lang.Cloneable

public class Factor

extends Predictor

A Factor is a Predictor that has discrete levels. This class provides for factors having either fixed or random levels.

Version:

2002.8.6

Author:

Russ Lenth

Fields inherited from class rvl.datasim.Predictor

ARRAY, component, COVARIATE, FACTOR, RARRAY, RESPONSE

Constructor Summary

Factor(java.lang.String name, int nLevels, double SD)
Construct a random factor named name with nLevels levels that will be labeled 1, 2, ..., nLevels.

Factor(java.lang.String name, java.lang.String[] levelNames, double SD)
Construct a random factor named name with levels named as specified by labels in levelNames.

Factor(java.lang.String name, java.lang.String[] levelNames, double[] effects)
Construct a fixed factor named name with levels named as specified by labels in levelNames and effects in effects

Method Summary

double	getEffect() This method should be overridden in derived classes.
double[]	getEffects()
int	getLevel()
java.lang.String	getLevelName()
java.lang.String[]	getLevelNames()
void	setAuto() Make the factor "automatic"; i.e., its level advances each time that data are collected.
void	setLevel(int level) set the current level of the factor to level
void	setLevels(java.lang.String[] lNames, double[] effs) Set names of factor level names and effects.
void	setRandomLevels() Set the effects based on a new set of normal random numbers with the standard deviation previously specified in construction.
void	setRandomLevels(double sd) Set the effects based on a new set of normal random numbers with standard deviation sd.

Methods inherited from class rvl.datasim.Predictor

clone, getName, getType, setVisible

Methods inherited from class java.lang.Object

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

Constructor Detail

Factor

public Factor(java.lang.String name,
              int nLevels,
              double SD)

Construct a random factor named name with nLevels levels that will be labeled 1, 2, ..., nLevels. At time of construction, the effects are generated from a normal distribution having mean zero and standard deviation SD.

Factor

public Factor(java.lang.String name,
              java.lang.String[] levelNames,
              double SD)

Construct a random factor named name with levels named as specified by labels in levelNames. At time of construction, the effects are generated from a normal distribution having mean zero and standard deviation SD.

Factor

public Factor(java.lang.String name,
              java.lang.String[] levelNames,
              double[] effects)

Construct a fixed factor named name with levels named as specified by labels in levelNames and effects in effects

Method Detail

getEffect

public double getEffect()

Description copied from class: Predictor

This method should be overridden in derived classes.

Overrides:

getEffect in class Predictor

Returns:

the effect at the current level

getEffects

public double[] getEffects()

Returns:

the factor effects

getLevel

public int getLevel()

Returns:

the current level

getLevelName

public java.lang.String getLevelName()

Returns:

the label associated with the current level

getLevelNames

public java.lang.String[] getLevelNames()

Returns:

the labels of the factor levels

setAuto

public void setAuto()

Make the factor "automatic"; i.e., its level advances each time that data are collected. If you do this, you probably also want to write a public int nextRun() method to advance the factor level(s)

setLevel

public void setLevel(int level)

set the current level of the factor to level

setLevels

public void setLevels(java.lang.String[] lNames,
                      double[] effs)

Set names of factor level names and effects. The two arrays must be the same length. Note: factor is reset to its first level (level 0).

setRandomLevels

public void setRandomLevels()

Set the effects based on a new set of normal random numbers with the standard deviation previously specified in construction.

setRandomLevels

public void setRandomLevels(double sd)

Set the effects based on a new set of normal random numbers with standard deviation sd.