com.numericalmethod.suanshu.stats.dlm
Class Dlm

java.lang.Object
  com.numericalmethod.suanshu.stats.dlm.Dlm

public class Dlm
extends java.lang.Object

This class represents a controlled DLM (controlled dynamic linear model) specification.

A controlled dynamic linear model (controlled DLM) is a generalization of the dynamic linear model (DLM) by including control variables. For t = 1, 2, ..., T, it has the following form:

y_t = F_t * x_t + v_t (Observation Equation),

x_t = G_t * x_{t - 1} + H_t * u_t + w_t (State Equation),

where the observations {y_t} and states {x_t} (t >= 1) are d-dimensional and p-dimensional vectors respectively; F_t and G_t are known matrices of dimension (d * p) and (p * p) respectively; {v_t} and {w_t} (t >= 1) are two independent sequences of independent normal random vectors with mean zero and known variance matrices {V_t} and {W_t} (t >= 1), respectively; Furthermore, it is assumed that x_0 is independent of {v_t} and {w_t} and is normally distributed with mean m_0 and covariance matrix C_0, where m_0 is a p-dimensional vector and C_0 is a (p * p) matrix; u_t is an m-dimensional vector of control variables, i.e., variables whose value can be regulated by the user, in order to obtain a desired level of the state x_t; H_t is a known (p * m) matrix of coefficients.

See Also:

• Wikipedia: Kalman filter - Underlying dynamic system model
• G. Petris et al., "Dynamic Linear Models with R," New York, Springer, 2009, ch. 2, pp. 31-84.

Constructor Summary

Dlm(Dlm that)
Copy constructor.

Dlm(Vector m0, Matrix C0, ObservationEquation Yt, StateEquation Xt)
Construct a (multivariate) controlled dynamic linear model.

Method Summary

ImmutableMatrix	C0() Get the covariance matrix of x_0.
int	d() Get the dimension of observations.
ObservationEquation	getObservationModel() Get the observation model.
StateEquation	getStateModel() Get the state model.
ImmutableVector	m0() Get the the mean of x_0.
int	p() Get the dimension of states.

Methods inherited from class java.lang.Object

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

Constructor Detail

Dlm

public Dlm(Vector m0,
           Matrix C0,
           ObservationEquation Yt,
           StateEquation Xt)

Construct a (multivariate) controlled dynamic linear model.

Parameters:

m0 - the mean of x_0

C0 - the covariance matrix of x_0

Yt - the observation equation for the model

Xt - the state equation for the model

Dlm

public Dlm(Dlm that)

Copy constructor.

Parameters:

that - a (multivariate) controlled dynamic linear model

Method Detail

m0

public ImmutableVector m0()

Get the the mean of x_0.

Returns:

m0, the mean of x_0

C0

public ImmutableMatrix C0()

Get the covariance matrix of x_0.

Returns:

C0, the covariance matrix of x_0

d

public int d()

Get the dimension of observations.

Returns:

d, the dimension of observations

p

public int p()

Get the dimension of states.

Returns:

p, the dimension of states

getObservationModel

public ObservationEquation getObservationModel()

Get the observation model.

Returns:

the observation model

getStateModel

public StateEquation getStateModel()

Get the state model.

Returns:

the state model