arorder: Algorithm for Order of an Autoregressive Time Series

In kloke/npsmReg2: R functions for Chapter 7 of Nonparametric Statistical Methods Using R

Description

This is an R implementation of the order of an autoregressive series proposed in Terpstra et al. (2001). The R function is discussed in detail in Section 7.8.1 of Kloke and McKean (2014). The sequential testing is a Wald-tyoe of test.

Usage

 arorder(n, maxp, est, varcov, alpha = 0.05) 

Arguments

n	length of series
maxp	maximal order
est	estimates of time series coefficients for AR(maxp)
varcov	variance-covariance matrix of estimates
alpha	significance level, (default is 0.05)

Value

orderofar	order selected by algorithm
results	matrix with stepwise results from algorithm

Author(s)

Joe McKean [e-mail]

References

Kloke, J. and McKean, J.W. (2014), Nonparametric statistical methods using R, Boca Raton, FL: Chapman-Hall.

Terpstra, J, McKean, J.W., and Naranjo, J.D. (2001), GR-estimates for an autoregressive time series, Statistics and Probability Letters, 51, 165-172.

Examples

data(ar2)
data <- lagmat(ar2,4)
x <- data[,1]
xmat <- data[,2:(4+1)]
hbrv<- hbrfit(x~xmat)
varcov <- vcov(hbrv,details=TRUE)
est<-hbrv$coef
alg<-arorder(length(x),4,est,varcov)

kloke/npsmReg2 documentation built on May 20, 2019, 12:34 p.m.