runAR: Running Autocorrelations

In kcpRS: Kernel Change Point Detection on the Running Statistics

Running Autocorrelations

Description

Extracts the running autocorrelations by sliding a window comprised of wsize time points, and in each window, the autocorrelation for each variable is computed. Each time the window is slid, the oldest time point is discarded and the latest time point is added.

Usage

runAR(data, wsize = 25)

Arguments

data	N x v dataframe where N is the no. of time points and v the no. of variables
wsize	Window size

Value

Running autocorrelations time series

Examples

phase1=cbind(rnorm(50,0,1),rnorm(50,0,1)) #phase1: AutoCorr=0
phase2=cbind(rnorm(50,0,1),rnorm(50,0,1))
phase2=filter(phase2,.50, method="recursive") #phase2: AutoCorr=.5
X=rbind(phase1,phase2)
RS=runAR(data=X,wsize=25)
ts.plot(RS, gpars=list(xlab="Window", ylab="Autocorrelation", col=1:2,lwd=2))

kcpRS documentation built on Oct. 25, 2023, 5:07 p.m.