kalman {hdiVAR}

kalman filtering and smoothing for vector autoregression with measurement error

Description

kalman filtering and smoothing for vector autoregression with measurement error

Usage

kalman(Y,A,sig_eta,sig_epsilon,X_init=NULL,P_init=NULL)

Arguments

Y

observations of time series, a p by T matrix.

A

current estimate of transition matrix.

sig_eta

current estiamte of \sigma_\eta.

sig_epsilon

current estiamte \sigma_\epsilon.

X_init

inital estimate of latent x_1 at the first iteration, a p-dimensional vector.

P_init

inital covariance estimate of latent x_1 at the first iteration, a p by p matrix.

Value

a list of conditional expectations and covariances of x_t's.

Author(s)

Xiang Lyu, Jian Kang, Lexin Li