com.numericalmethod.suanshu.stats.timeseries.multivariate

Class GenericTimeTimeSeries<T extends java.lang.Comparable<? super T>>

java.lang.Object

com.numericalmethod.suanshu.stats.timeseries.multivariate.GenericTimeTimeSeries<T>

Type Parameters:

T - the timestamp type

All Implemented Interfaces:

MultiVariateTimeSeries<T,MultiVariateTimeSeries.Entry<T>>, TimeSeries<T,Vector,MultiVariateTimeSeries.Entry<T>>, java.io.Serializable, java.lang.Iterable<MultiVariateTimeSeries.Entry<T>>

public class GenericTimeTimeSeries<T extends java.lang.Comparable<? super T>>
extends java.lang.Object
implements MultiVariateTimeSeries<T,MultiVariateTimeSeries.Entry<T>>

This is a multivariate time series indexed by some notion of time. This implementation explicitly remembers all data in memory. It is appropriate for a short time series.

See Also:

Wikipedia: Time series, Serialized Form

Nested Class Summary

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

MultiVariateTimeSeries.Entry<T>

Constructor Summary

Constructors

Constructor and Description
GenericTimeTimeSeries(T[] timestamps, double[][] values) Construct a multivariate time series from timestamps and vectors.
GenericTimeTimeSeries(T[] timestamps, Matrix values) Construct a multivariate time series from timestamps and vectors.
GenericTimeTimeSeries(T[] timestamps, Vector[] values) Construct a multivariate time series from timestamps and vectors.

Method Summary

Modifier and Type	Method and Description
GenericTimeTimeSeries<T>	diff(int d) Construct an instance of GenericTimeTimeSeries by taking the first difference d times.
int	dimension() Get the dimension of the multivariate time series.
GenericTimeTimeSeries<T>	drop(int nItems) Construct an instance of GenericTimeTimeSeries by dropping the leading nItems entries, those most backward in time entries.
boolean	equals(java.lang.Object obj)
Vector	get(int i) Get the i-th value.
int	hashCode()
java.util.Iterator<MultiVariateTimeSeries.Entry<T>>	iterator()
int	size() Get the length of the time series.
T	time(int i) Get the i-th timestamp.
T[]	timestamps() Get all the timestamps.
Vector[]	toArray() Convert this multivariate time series into an array of vectors.
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

Methods inherited from interface java.lang.Iterable

forEach, spliterator

Constructor Detail

GenericTimeTimeSeries

public GenericTimeTimeSeries(T[] timestamps,
                             Vector[] values)

Construct a multivariate time series from timestamps and vectors.

Parameters:

timestamps - the timestamps

values - the vector values

GenericTimeTimeSeries

public GenericTimeTimeSeries(T[] timestamps,
                             double[][] values)

Construct a multivariate time series from timestamps and vectors.

Parameters:

timestamps - the timestamps

values - the vector values

GenericTimeTimeSeries

public GenericTimeTimeSeries(T[] timestamps,
                             Matrix values)

Construct a multivariate time series from timestamps and vectors.

Parameters:

timestamps - the timestamps

values - the vector values

Method Detail

dimension

public int dimension()

Description copied from interface: MultiVariateTimeSeries

Get the dimension of the multivariate time series.

Specified by:

dimension in interface MultiVariateTimeSeries<T extends java.lang.Comparable<? super T>,MultiVariateTimeSeries.Entry<T extends java.lang.Comparable<? super T>>>

Returns:

the multivariate time series dimension

size

public int size()

Description copied from interface: TimeSeries

Get the length of the time series.

Specified by:

size in interface TimeSeries<T extends java.lang.Comparable<? super T>,Vector,MultiVariateTimeSeries.Entry<T extends java.lang.Comparable<? super T>>>

Returns:

the time series length

iterator

public java.util.Iterator<MultiVariateTimeSeries.Entry<T>> iterator()

Specified by:

iterator in interface java.lang.Iterable<MultiVariateTimeSeries.Entry<T extends java.lang.Comparable<? super T>>>

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<T extends java.lang.Comparable<? super T>,MultiVariateTimeSeries.Entry<T extends java.lang.Comparable<? super T>>>

Returns:

the matrix representation of this time series

toArray

public Vector[] toArray()

Convert this multivariate time series into an array of vectors.

Returns:

the array representation of this time series

get

public Vector get(int i)

Get the i-th value.

Parameters:

i - the position index

Returns:

the i-th value

time

public T time(int i)

Get the i-th timestamp.

Parameters:

i - the position index

Returns:

the i-th timestamp

timestamps

public T[] timestamps()

Get all the timestamps.

Returns:

the timestamps

drop

public GenericTimeTimeSeries<T> drop(int nItems)

Construct an instance of GenericTimeTimeSeries by dropping the leading nItems entries, those most backward in time entries.

Parameters:

nItems - the number of leading entries to be dropped

Returns:

an instance of GenericTimeTimeSeries

diff

public GenericTimeTimeSeries<T> diff(int d)

Construct an instance of GenericTimeTimeSeries by taking the first difference d times.

Parameters:

d - the number of differences

Returns:

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

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