com.numericalmethod.suanshu.stats.timeseries.multivariate.realtime
Class SimpleMultiVariateTimeSeries

java.lang.Object
  com.numericalmethod.suanshu.stats.timeseries.multivariate.realtime.SimpleMultiVariateTimeSeries

All Implemented Interfaces:

MultiVariateTimeSeries<java.lang.Integer>, TimeSeries<java.lang.Integer,Vector>

Direct Known Subclasses:

ArimaSim

public class SimpleMultiVariateTimeSeries
extends java.lang.Object
implements MultiVariateTimeSeries

A simple multivariate time series has its vectored values indexed by integers. Data are assumed to be equally spaced.

This time of series is often seen in classical multivariate time series analysis.

Nested Class Summary

class

SimpleMultiVariateTimeSeries.Iterator the Iterator to read a SimpleMultiVariateTimeSeries

Nested classes/interfaces inherited from interface com.numericalmethod.suanshu.stats.timeseries.multivariate.realtime.MultiVariateTimeSeries

MultiVariateTimeSeries.Entry

Constructor Summary

SimpleMultiVariateTimeSeries(double[]... values)
Construct a SimpleMultiVariateTimeSeries.

SimpleMultiVariateTimeSeries(Matrix values)
Construct a SimpleMultiVariateTimeSeries.

SimpleMultiVariateTimeSeries(TimeSeries ts)
Construct a SimpleMultiVariateTimeSeries from a univariate time series.

SimpleMultiVariateTimeSeries(Vector... values)
Construct a SimpleMultiVariateTimeSeries.

Method Summary

SimpleMultiVariateTimeSeries	diff(int d) Construct a new and independent SimpleMultiVariateTimeSeries by taking the first difference.
int	dimension() Get the dimensionality of the multivariate time series.
SimpleMultiVariateTimeSeries	drop(int nItems) Construct a new and independent SimpleMultiVariateTimeSeries by dropping the leading nItems entries.
boolean	equals(java.lang.Object obj)
Vector	get(int t) Get the value at time t.
int	hashCode()
SimpleMultiVariateTimeSeries.Iterator	iterator() Get an Iterator to read this integer-indexed multivariate time series.
SimpleMultiVariateTimeSeries	lag(int nLags) Construct a new and independent SimpleMultiVariateTimeSeries by lagging the time series.
SimpleMultiVariateTimeSeries	lag(int nLags, int length) Construct a new and independent SimpleMultiVariateTimeSeries by lagging the time series.
int	size() the length of the time series
Matrix	toMatrix() Convert this multivariate time series into an m x n matrix, where m is the dimension, and n the length.
java.lang.String	toString()

Methods inherited from class java.lang.Object

clone, finalize, getClass, notify, notifyAll, wait, wait, wait

Constructor Detail

SimpleMultiVariateTimeSeries

public SimpleMultiVariateTimeSeries(Matrix values)

Construct a SimpleMultiVariateTimeSeries.

Parameters:

values - a matrix representation of a time series

SimpleMultiVariateTimeSeries

public SimpleMultiVariateTimeSeries(double[]... values)

Construct a SimpleMultiVariateTimeSeries.

Parameters:

values - a double array representation of a time series

SimpleMultiVariateTimeSeries

public SimpleMultiVariateTimeSeries(Vector... values)

Construct a SimpleMultiVariateTimeSeries.

Parameters:

values - a vector representation of a time series

SimpleMultiVariateTimeSeries

public SimpleMultiVariateTimeSeries(TimeSeries ts)

Construct a SimpleMultiVariateTimeSeries from a univariate time series.

Parameters:

ts - a univariate time series

Method Detail

iterator

public SimpleMultiVariateTimeSeries.Iterator iterator()

Description copied from interface: MultiVariateTimeSeries

Get an Iterator to read this integer-indexed multivariate time series.

Specified by:

iterator in interface MultiVariateTimeSeries<java.lang.Integer>

Specified by:

iterator in interface TimeSeries<java.lang.Integer,Vector>

Returns:

an iterator

toMatrix

public Matrix toMatrix()

Description copied from interface: MultiVariateTimeSeries

Convert this multivariate time series into an m x n matrix, where m is the dimension, and n the length.

Specified by:

toMatrix in interface MultiVariateTimeSeries<java.lang.Integer>

Returns:

the matrix representation of this time series

size

public int size()

Description copied from interface: TimeSeries

the length of the time series

Specified by:

size in interface TimeSeries<java.lang.Integer,Vector>

Returns:

the size

dimension

public int dimension()

Description copied from interface: MultiVariateTimeSeries

Get the dimensionality of the multivariate time series.

Specified by:

dimension in interface MultiVariateTimeSeries<java.lang.Integer>

Returns:

the dimensionality

get

public Vector get(int t)

Description copied from interface: MultiVariateTimeSeries

Get the value at time t.

Parameters:

t - the time index, ranging from 1 up to T

Returns:

ts[t]

drop

public SimpleMultiVariateTimeSeries drop(int nItems)

Construct a new and independent SimpleMultiVariateTimeSeries by dropping the leading nItems entries.

Parameters:

nItems - the number of leading entries to be dropped

Returns:

a new and independent time series

diff

public SimpleMultiVariateTimeSeries diff(int d)

Construct a new and independent SimpleMultiVariateTimeSeries by taking the first difference.

Parameters:

d - the number of differences

Returns:

diff(x, lag = 1, differences = d) as in R

lag

public SimpleMultiVariateTimeSeries lag(int nLags,
                                        int length)

Construct a new and independent SimpleMultiVariateTimeSeries by lagging the time series.

This only makes sense for equi-distant data points.

Parameters:

nLags - the number of lags

length - the length of the lagged time series

Returns:

a lagged time series

lag

public SimpleMultiVariateTimeSeries lag(int nLags)

Construct a new and independent SimpleMultiVariateTimeSeries by lagging the time series.

This only makes sense for equi-distant data points.

Parameters:

nLags - the number of lags

Returns:

a lagged time series

toString

public java.lang.String toString()

Overrides:

toString in class java.lang.Object

equals

public boolean equals(java.lang.Object obj)

Overrides:

equals in class java.lang.Object

hashCode

public int hashCode()

Overrides:

hashCode in class java.lang.Object