marfit: Yule-Walker Method of Fitting Multivariate AR Model

In TSSS: Time Series Analysis with State Space Model

Yule-Walker Method of Fitting Multivariate AR Model

Description

Fit a multivariate AR model by the Yule-Walker method.

Usage

marfit(y, lag = NULL)

Arguments

y	a multivariate time series.
lag	highest order of fitted AR models. Default is 2 \sqrt{n}, where n is the length of the time series y.

Value

An object of class "maryule", which is a list with the following components:

maice.order	order of minimum AIC.
aic	AIC's of the AR models with order 0,\dots,lag.
v	innovation covariance matrix of the AIC best model.
arcoef	AR coefficients of the AIC best model.

References

Kitagawa, G. (2020) Introduction to Time Series Modeling with Applications in R. Chapman & Hall/CRC.

Examples

# Yaw rate, rolling, pitching and rudder angle of a ship
data(HAKUSAN)
yy <- as.matrix(HAKUSAN[, c(1,2,4)])   # Yaw rate, Pitching, Rudder angle
nc <- dim(yy)[1]
n <- seq(1, nc, by = 2) 
y <- yy[n, ]
marfit(y, 20)

TSSS documentation built on Sept. 29, 2023, 9:07 a.m.