HoltWinters (timeseriesForecasting 1.1.27 API)

java.lang.Object
- weka.classifiers.AbstractClassifier
- - weka.classifiers.timeseries.HoltWinters

All Implemented Interfaces:

Serializable, Cloneable, weka.classifiers.Classifier, PrimingDataLearner, weka.core.BatchPredictor, weka.core.CapabilitiesHandler, weka.core.CapabilitiesIgnorer, weka.core.CommandlineRunnable, weka.core.OptionHandler, weka.core.RevisionHandler
```
public class HoltWinters
extends weka.classifiers.AbstractClassifier
implements PrimingDataLearner, weka.core.OptionHandler, Serializable
```
Class implementing the Holt-Winters triple exponential smoothing method for time series forecasting. Designed to be used in the Weka forecasting environment. Although lagged variables are not used by this method, for evaluation purposes within the framework the user is required to set a lag length that is at least 3 times that of the seasonal cycle length. This is because priming data (which is set by the framework to have as many instances as the lag length) is used to train the Holt-Winters forecaster. Two seasonal cycles worth of data are used to set the initial parameters of the method (primarily the seasonal components) and the remaining training data is used to smooth over.

Version:

$Revision: $

Author:

Mark Hall (mhall{[at]}pentaho{[dot]}com)

See Also:
Serialized Form

Field Summary
- Fields inherited from class weka.classifiers.AbstractClassifier
  BATCH_SIZE_DEFAULT, NUM_DECIMAL_PLACES_DEFAULT

Constructor Summary

Constructors
Constructor and Description

HoltWinters()

Constructors
Constructor and Description
`HoltWinters()`

Method Summary

Methods
Modifier and Type	Method and Description
`void`	`buildClassifier(weka.core.Instances data)`
`double`	`classifyInstance(weka.core.Instance inst)`
`String`	`excludeSeasonalCorrectionTipText()` Tip text for this property
`String`	`excludeTrendCorrection()` Tip text for this property
`double`	`forecast()` Generates a one-step ahead forecast.
`boolean`	`getExcludeSeasonalCorrection()` Get whether to exclude the seasonal correction
`boolean`	`getExcludeTrendCorrection()` Get whether to exclude the trend correction
`int`	`getMinRequiredTrainingPoints()` Return the minimum number of training/priming data points required before a forecast can be made
`String[]`	`getOptions()`
`double`	`getSeasonalSmoothingFactor()` Get the seasonal smoothing factor
`int`	`getSeasonCycleLength()` Get the length of a "year", i.e.
`double`	`getTrendSmoothingFactor()` Get the trend smoothing factor
`double`	`getValueSmoothingFactor()` Get the value smoothing factor
`String`	`globalInfo()` Description of this forecaster.
`Enumeration<weka.core.Option>`	`listOptions()`
`static void`	`main(String[] args)` Main method for running this class
`static void`	`printUsage()` Print usage information
`void`	`reset()` Reset this forecaster ready to learn from a new set of priming data
`String`	`seasonalSmoothingFactorTipText()` Tip text for this property
`String`	`seasonCycleLengthTipText()` Tip text for this property
`void`	`setExcludeSeasonalCorrection(boolean s)` Set whether to exclude the seasonal correction
`void`	`setExcludeTrendCorrection(boolean t)` Set whether to exclude the trend correction
`void`	`setOptions(String[] options)`
`void`	`setSeasonalSmoothingFactor(double s)` Set the seasonal smoothing factor
`void`	`setSeasonCycleLength(int length)` Set the length of a "year", i.e.
`void`	`setTrendSmoothingFactor(double t)` Set the trend smoothing factor
`void`	`setValueSmoothingFactor(double s)` Set the value smoothing factor
`String`	`toString()`
`String`	`trendSmoothingFactorTipText()` Tip text for this property
`void`	`updateForecaster(double primingOrPredictedTargetValue)` Update the smoothed estimates using the supplied value
`String`	`valueSmoothingFactorTipText()` Tip text for this property

Methods inherited from class weka.classifiers.AbstractClassifier
batchSizeTipText, debugTipText, distributionForInstance, distributionsForInstances, doNotCheckCapabilitiesTipText, forName, getBatchSize, getCapabilities, getDebug, getDoNotCheckCapabilities, getNumDecimalPlaces, getRevision, implementsMoreEfficientBatchPrediction, makeCopies, makeCopy, numDecimalPlacesTipText, postExecution, preExecution, run, runClassifier, setBatchSize, setDebug, setDoNotCheckCapabilities, setNumDecimalPlaces

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

- Constructor Detail
  - HoltWinters
```
public HoltWinters()
```
- Method Detail
  - globalInfo
```
public String globalInfo()
```
    Description of this forecaster.
    
    Returns:
    a description of this forecaster
  - reset
```
public void reset()
```
    Description copied from interface: PrimingDataLearner
    
    Reset this forecaster ready to learn from a new set of priming data
    
    Specified by:
    
    reset in interface PrimingDataLearner
  - getMinRequiredTrainingPoints
```
public int getMinRequiredTrainingPoints()
```
    Return the minimum number of training/priming data points required before a forecast can be made
    
    Specified by:
    
    getMinRequiredTrainingPoints in interface PrimingDataLearner
    
    Returns:
    the minimum number of training/priming data points required
  - updateForecaster
```
public void updateForecaster(double primingOrPredictedTargetValue)
```
    Update the smoothed estimates using the supplied value
    
    Specified by:
    
    updateForecaster in interface PrimingDataLearner
    
    Parameters:
    primingOrPredictedTargetValue - the value to update with
  - forecast
```
public double forecast()
                throws Exception
```
    Generates a one-step ahead forecast. Clients should follow this with a call to updateForecaster() and pass in the forecasted value if they want to generate further projected values (i.e. beyond one-step ahead).
    
    Returns:
    a one-step ahead forecast
    
    Throws:
    
    Exception - if forecast can't be produced
  - buildClassifier
```
public void buildClassifier(weka.core.Instances data)
                     throws Exception
```
    Specified by:
    
    buildClassifier in interface weka.classifiers.Classifier
    
    Throws:
    
    Exception
  - classifyInstance
```
public double classifyInstance(weka.core.Instance inst)
                        throws Exception
```
    Specified by:
    
    classifyInstance in interface weka.classifiers.Classifier
    
    Overrides:
    
    classifyInstance in class weka.classifiers.AbstractClassifier
    
    Throws:
    
    Exception
  - listOptions
```
public Enumeration<weka.core.Option> listOptions()
```
    Specified by:
    
    listOptions in interface weka.core.OptionHandler
    
    Overrides:
    
    listOptions in class weka.classifiers.AbstractClassifier
  - setOptions
```
public void setOptions(String[] options)
                throws Exception
```
    Specified by:
    
    setOptions in interface weka.core.OptionHandler
    
    Overrides:
    
    setOptions in class weka.classifiers.AbstractClassifier
    
    Throws:
    
    Exception
  - getOptions
```
public String[] getOptions()
```
    Specified by:
    
    getOptions in interface weka.core.OptionHandler
    
    Overrides:
    
    getOptions in class weka.classifiers.AbstractClassifier
  - seasonCycleLengthTipText
```
public String seasonCycleLengthTipText()
```
    Tip text for this property
    
    Returns:
    the tip text for this property
  - setSeasonCycleLength
```
public void setSeasonCycleLength(int length)
```
    Set the length of a "year", i.e. the number of seasons for the seasonal cycle
    
    Parameters:
    length - the number of seasons in a cycle
  - getSeasonCycleLength
```
public int getSeasonCycleLength()
```
    Get the length of a "year", i.e. the number of seasons for the seasonal cycle
    
    Parameters:
    length - the number of seasons in a cycle
  - excludeSeasonalCorrectionTipText
```
public String excludeSeasonalCorrectionTipText()
```
    Tip text for this property
    
    Returns:
    the tip text for this property
  - setExcludeSeasonalCorrection
```
public void setExcludeSeasonalCorrection(boolean s)
```
    Set whether to exclude the seasonal correction
    
    Parameters:
    s - true if the seasonal correction is to be excluded
  - getExcludeSeasonalCorrection
```
public boolean getExcludeSeasonalCorrection()
```
    Get whether to exclude the seasonal correction
    
    Returns:
    true if the seasonal correction is to be excluded
  - excludeTrendCorrection
```
public String excludeTrendCorrection()
```
    Tip text for this property
    
    Returns:
    the tip text for this property
  - setExcludeTrendCorrection
```
public void setExcludeTrendCorrection(boolean t)
```
    Set whether to exclude the trend correction
    
    Parameters:
    s - true if the trend correction is to be excluded
  - getExcludeTrendCorrection
```
public boolean getExcludeTrendCorrection()
```
    Get whether to exclude the trend correction
    
    Returns:
    true if the trend correction is to be excluded
  - valueSmoothingFactorTipText
```
public String valueSmoothingFactorTipText()
```
    Tip text for this property
    
    Returns:
    the tip text for this property
  - setValueSmoothingFactor
```
public void setValueSmoothingFactor(double s)
```
    Set the value smoothing factor
    
    Parameters:
    s - the value smoothing factor
  - getValueSmoothingFactor
```
public double getValueSmoothingFactor()
```
    Get the value smoothing factor
    
    Returns:
    the value smoothing factor
  - trendSmoothingFactorTipText
```
public String trendSmoothingFactorTipText()
```
    Tip text for this property
    
    Returns:
    the tip text for this property
  - setTrendSmoothingFactor
```
public void setTrendSmoothingFactor(double t)
```
    Set the trend smoothing factor
    
    Parameters:
    t - the trend smoothing factor
  - getTrendSmoothingFactor
```
public double getTrendSmoothingFactor()
```
    Get the trend smoothing factor
    
    Parameters:
    t - the trend smoothing factor
  - seasonalSmoothingFactorTipText
```
public String seasonalSmoothingFactorTipText()
```
    Tip text for this property
    
    Returns:
    the tip text for this property
  - setSeasonalSmoothingFactor
```
public void setSeasonalSmoothingFactor(double s)
```
    Set the seasonal smoothing factor
    
    Parameters:
    s - the seasonal smoothing factor
  - getSeasonalSmoothingFactor
```
public double getSeasonalSmoothingFactor()
```
    Get the seasonal smoothing factor
    
    Returns:
    the seasonal smoothing factor
  - printUsage
```
public static void printUsage()
```
    Print usage information
  - toString
```
public String toString()
```
    Overrides:
    
    toString in class Object
  - main
```
public static void main(String[] args)
```
    Main method for running this class
    
    Parameters:
    args - command line arguments

Class HoltWinters

Field Summary

Fields inherited from class weka.classifiers.AbstractClassifier

Constructor Summary

Method Summary

Methods inherited from class weka.classifiers.AbstractClassifier

Methods inherited from class java.lang.Object

Constructor Detail

HoltWinters

Method Detail

globalInfo

reset

getMinRequiredTrainingPoints

updateForecaster

forecast

buildClassifier

classifyInstance

listOptions

setOptions

getOptions

seasonCycleLengthTipText

setSeasonCycleLength

getSeasonCycleLength

excludeSeasonalCorrectionTipText

setExcludeSeasonalCorrection

getExcludeSeasonalCorrection

excludeTrendCorrection

setExcludeTrendCorrection

getExcludeTrendCorrection

valueSmoothingFactorTipText

setValueSmoothingFactor

getValueSmoothingFactor

trendSmoothingFactorTipText

setTrendSmoothingFactor

getTrendSmoothingFactor

seasonalSmoothingFactorTipText

setSeasonalSmoothingFactor

getSeasonalSmoothingFactor

printUsage

toString

main