org.hawkular.datamining.forecast.models

Interface TimeSeriesModel

All Known Implementing Classes:

AbstractExponentialSmoothing, ContinuousModel, DoubleExponentialSmoothing, SimpleExponentialSmoothing, TripleExponentialSmoothing

public interface TimeSeriesModel

Basic interface for time series model

All learn methods collects one step ahead prediction error runStatistics().

Author:

Pavol Loffay

Nested Class Summary

Nested Classes

Modifier and Type	Interface and Description
static class	TimeSeriesModel.TimestampComparator

Method Summary

Modifier and Type	Method and Description
DataPoint	forecast() One step ahead prediction
List<DataPoint>	forecast(int nAhead) Multi step ahead prediction
AccuracyStatistics	init(List<DataPoint> learnData) Initialize model and return statistics - error of one step ahead prediction
AccuracyStatistics	initStatistics()
long	lastTimestamp()
void	learn(DataPoint learnData) Learn one point
void	learn(List<DataPoint> learnData) Learn multiple points
int	minimumInitSize()
String	name()
int	numberOfParams()
AccuracyStatistics	runStatistics()

Method Detail

init

AccuracyStatistics init(List<DataPoint> learnData)

Initialize model and return statistics - error of one step ahead prediction

Parameters:

learnData - points to learn

Returns:

statistics for learnData

learn

void learn(DataPoint learnData)

Learn one point

Parameters:

learnData - point to learn

learn

void learn(List<DataPoint> learnData)

Learn multiple points

Parameters:

learnData - points to learn

forecast

DataPoint forecast()

One step ahead prediction

Returns:

predicted point

forecast

List<DataPoint> forecast(int nAhead)

Multi step ahead prediction

Parameters:

nAhead - number of steps for forecasting

Returns:

predicted points

initStatistics

AccuracyStatistics initStatistics()

Returns:

initial statistics calculated when calling initStatistics()

runStatistics

AccuracyStatistics runStatistics()

Returns:

statistics calculated on learning data

name

String name()

Returns:

name of the implemented model

numberOfParams

int numberOfParams()

Returns:

number of parameters of the model, seasonal models include seasonal indices

minimumInitSize

int minimumInitSize()

Returns:

number of parameters of the model, seasonal models include seasonal indices

lastTimestamp

long lastTimestamp()