PELT.MVar: Most recent changepoints from MV method using censored AR...

In cpcens: Changepoint Analysis using Censored Time Series Data

Description

Detecting most recent changepoints from MV methd (Lavielle and Teyssiere, 2006) deal with multivariate data which is modeling the data within each segment as a multivariate (MV) Gaussian having a given covariance after generating censored data from AR model.

Usage

PELT.MVar(data, beta = 101 * log(dim(data)[2]))

Arguments

data	a censored data matrix obtained from AR1.data .
beta	default 101*log(dim(data)[2])). Here dim(data)[2] means consider size(length) of series (n).

Value

indicates the most recent changepoint in each series .

References

Lavielle, M. and Teyssiere, G. (2006). Detection of multiple changepoints in multivariate time series.Lithuanian Mathematical Journal, 46(3):287-306

See Also

AR1.data

Examples

# example (Right censoring)
library(cpcens)
# The size of series(n) should be greater than 200.
sim=AR1.data(n = 500, N = 100, K = 5, eps = 1,
rho = 0.6, mu = 0, siga = 1, rates = c(NA, 0.2), Mrate = 0)
data=sim$data
N=100
pmv = PELT.MVar( data , 101*log(dim(data)[2]) )
mv.chpts =  rep( rev( pmv$cpts )[1] , N )

cpcens documentation built on Aug. 2, 2019, 5:05 p.m.