UnivariateKernelEstimator (interruptible-weka 0.1.2 API)

java.lang.Object
- weka.estimators.UnivariateKernelEstimator

All Implemented Interfaces:

java.io.Serializable, RevisionHandler, UnivariateDensityEstimator, UnivariateIntervalEstimator, UnivariateQuantileEstimator
```
public class UnivariateKernelEstimator
extends java.lang.Object
implements UnivariateDensityEstimator, UnivariateIntervalEstimator, UnivariateQuantileEstimator, java.io.Serializable
```
Simple weighted kernel density estimator.

See Also:

Serialized Form

Field Summary

Fields
Modifier and Type	Field and Description
`static double`	`CONST` Constant for Gaussian density.
`protected double`	`m_Exponent` The exponent to use in computation of bandwidth (default: -0.25)
`protected double`	`m_MinWidth` The minimum allowed value of the kernel width (default: 1.0E-6)
`protected int`	`m_NumIntervals` The number of intervals used to approximate prediction interval.
`protected double`	`m_SumOfWeights` The weight of the values collected so far
`protected double`	`m_Threshold` Threshold at which further kernels are no longer added to sum.
`protected java.util.TreeMap<java.lang.Double,java.lang.Double>`	`m_TM` The collection used to store the weighted values.
`protected double`	`m_WeightedSum` The weighted sum of values
`protected double`	`m_WeightedSumSquared` The weighted sum of squared values
`protected double`	`m_Width` The current bandwidth (only computed when needed)

Constructor Summary

Constructors
Constructor and Description

UnivariateKernelEstimator()

Constructors
Constructor and Description
`UnivariateKernelEstimator()`

Method Summary

All Methods Static Methods Instance Methods Concrete Methods
Modifier and Type	Method and Description
`void`	`addValue(double value, double weight)` Adds a value to the density estimator.
`java.lang.String`	`getRevision()` Returns the revision string.
`java.lang.String`	`globalInfo()` Returns a string describing the estimator.
`double`	`logDensity(double value)` Returns the natural logarithm of the density estimate at the given point.
`protected double`	`logOfSum(double logOfX, double logOfY)` Computes the logarithm of x and y given the logarithms of x and y.
`static void`	`main(java.lang.String[] args)` Main method, used for testing this class.
`double[][]`	`predictIntervals(double conf)` Returns the interval for the given confidence value.
`double`	`predictQuantile(double percentage)` Returns the quantile for the given percentage.
`protected void`	`runningSum(java.util.Set<java.util.Map.Entry<java.lang.Double,java.lang.Double>> c, double value, double[] sums)` Compute running sum of density values and weights.
`java.lang.String`	`toString()` Returns textual description of this estimator.
`void`	`updateWidth()` Updates bandwidth: the sample standard deviation is multiplied by the total weight to the power of the given exponent.

Methods inherited from class java.lang.Object
clone, equals, finalize, getClass, hashCode, notify, notifyAll, wait, wait, wait

- Field Detail
  - m_TM
```
protected java.util.TreeMap<java.lang.Double,java.lang.Double> m_TM
```
    The collection used to store the weighted values.
  - m_WeightedSum
```
protected double m_WeightedSum
```
    The weighted sum of values
  - m_WeightedSumSquared
```
protected double m_WeightedSumSquared
```
    The weighted sum of squared values
  - m_SumOfWeights
```
protected double m_SumOfWeights
```
    The weight of the values collected so far
  - m_Width
```
protected double m_Width
```
    The current bandwidth (only computed when needed)
  - m_Exponent
```
protected double m_Exponent
```
    The exponent to use in computation of bandwidth (default: -0.25)
  - m_MinWidth
```
protected double m_MinWidth
```
    The minimum allowed value of the kernel width (default: 1.0E-6)
  - CONST
```
public static final double CONST
```
    Constant for Gaussian density.
  - m_Threshold
```
protected double m_Threshold
```
    Threshold at which further kernels are no longer added to sum.
  - m_NumIntervals
```
protected int m_NumIntervals
```
    The number of intervals used to approximate prediction interval.
- Constructor Detail
  - UnivariateKernelEstimator
```
public UnivariateKernelEstimator()
```
- Method Detail
  - globalInfo
```
public java.lang.String globalInfo()
```
    Returns a string describing the estimator.
  - addValue
```
public void addValue(double value,
                     double weight)
```
    Adds a value to the density estimator.
    
    Specified by:
    
    addValue in interface UnivariateDensityEstimator
    
    Specified by:
    
    addValue in interface UnivariateIntervalEstimator
    
    Specified by:
    
    addValue in interface UnivariateQuantileEstimator
    
    Parameters:
    
    value - the value to add
    
    weight - the weight of the value
  - updateWidth
```
public void updateWidth()
```
    Updates bandwidth: the sample standard deviation is multiplied by the total weight to the power of the given exponent. If the total weight is not greater than zero, the width is set to Double.MAX_VALUE. If that is not the case, but the width becomes smaller than m_MinWidth, the width is set to the value of m_MinWidth.
  - predictIntervals
```
public double[][] predictIntervals(double conf)
```
    Returns the interval for the given confidence value.
    
    Specified by:
    
    predictIntervals in interface UnivariateIntervalEstimator
    
    Parameters:
    
    conf - the confidence value in the interval [0, 1]
    
    Returns:
    
    the interval
  - predictQuantile
```
public double predictQuantile(double percentage)
```
    Returns the quantile for the given percentage.
    
    Specified by:
    
    predictQuantile in interface UnivariateQuantileEstimator
    
    Parameters:
    
    percentage - the percentage
    
    Returns:
    
    the quantile
  - logOfSum
```
protected double logOfSum(double logOfX,
                          double logOfY)
```
    Computes the logarithm of x and y given the logarithms of x and y. This is based on Tobias P. Mann's description in "Numerically Stable Hidden Markov Implementation" (2006).
  - runningSum
```
protected void runningSum(java.util.Set<java.util.Map.Entry<java.lang.Double,java.lang.Double>> c,
                          double value,
                          double[] sums)
```
    Compute running sum of density values and weights.
  - logDensity
```
public double logDensity(double value)
```
    Returns the natural logarithm of the density estimate at the given point.
    
    Specified by:
    
    logDensity in interface UnivariateDensityEstimator
    
    Parameters:
    
    value - the value at which to evaluate
    
    Returns:
    
    the natural logarithm of the density estimate at the given value
  - toString
```
public java.lang.String toString()
```
    Returns textual description of this estimator.
    
    Overrides:
    
    toString in class java.lang.Object
  - getRevision
```
public java.lang.String getRevision()
```
    Returns the revision string.
    
    Specified by:
    
    getRevision in interface RevisionHandler
    
    Returns:
    
    the revision
  - main
```
public static void main(java.lang.String[] args)
```
    Main method, used for testing this class.

Class UnivariateKernelEstimator

Field Summary

Constructor Summary

Method Summary

Methods inherited from class java.lang.Object

Field Detail

m_TM

m_WeightedSum

m_WeightedSumSquared

m_SumOfWeights

m_Width

m_Exponent

m_MinWidth

CONST

m_Threshold

m_NumIntervals

Constructor Detail

UnivariateKernelEstimator

Method Detail

globalInfo

addValue

updateWidth

predictIntervals

predictQuantile

logOfSum

runningSum

logDensity

toString

getRevision

main