com.github.brandtg

Class StlDecomposition

java.lang.Object

com.github.brandtg.StlDecomposition

public class StlDecomposition
extends Object

Implementation of STL: A Seasonal-Trend Decomposition Procedure based on Loess.

Robert B. Cleveland et al., "STL: A Seasonal-Trend Decomposition Procedure based on Loess," in Journal of Official Statistics Vol. 6 No. 1, 1990, pp. 3-73

Author:

Greg Brandt, Jieying Chen, James Hong

Constructor Summary

Constructors

Constructor and Description
StlDecomposition(StlConfig config) Constructs a configuration of STL function that can de-trend data.

Method Summary

Modifier and Type	Method and Description
StlResult	decompose(double[] times, double[] series) Constructs the STL decomposition of a time series.

Methods inherited from class java.lang.Object

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

Constructor Detail

StlDecomposition

public StlDecomposition(StlConfig config)

Constructs a configuration of STL function that can de-trend data.

n.b. The Java Loess implementation only does linear local polynomial regression, but R supports linear (degree=1), quadratic (degree=2), and a strange degree=0 option.

Also, the Java Loess implementation accepts "bandwidth", the fraction of source points closest to the current point, as opposed to integral values.

Method Detail

decompose

public StlResult decompose(double[] times,
                           double[] series)

Constructs the STL decomposition of a time series.

times.length == series.length