cpcens (version 0.1.0)

multiple.mrc: Most recent changepoints from mrc method using censored AR/MA timeseries.

Description

Detecting most recent changepoints from mrc method consisting of many related univariate timeseries (Bardwell, Eckley, Fearnhead, and Smith, 2016) after generating censored data from AR/MA model and pools information across the time-series by solving the K-median problem using tb.raw (Teitz and Bart, 1968).

Usage

multiple.mrc(mrc, pmax = 10, alpha = 2, elbow.thresh = 0.5,
  n = 500)

Arguments

mrc

data obtained from mrc.mean

pmax

Maximum number of most recent changepoints to search for. Default value pmax=10.

alpha

The variable specific penalty used to penalise the addition of a given changepoint into a given variable. Default value alpha = 2.

elbow.thresh

default 0.5.

n

length of series

Value

indicates the most recent changepoint in each series .

References

Teitz, M. B. and Bart, P. (1968). Heuristic methods for estimating the generalized vertex median of a weighted graph. Operations Research, 16(5):955<U+2013>961.

Bardwell, L., Fearnhead, P., Eckley, I. A., Smith, S., and Spott, M. (2019). Most recent changepoint detection in panel data. Technometrics, 61(1):88<U+2013>98.

See Also

mrc

Examples

# NOT RUN {
#'#example(left censoring)
library(cpcens)
n=300
N=100
# Generate censored data using MA model
sim=MA1.data(n = 300, N = 100, K = 5, eps = 1,
rho = 0.4, mu = 0, siga = 1, rates = c(0.6,NA), Mrate = 0)
data=sim$data
mrc = mrc.mean( data , beta = 1.5*log(n) )
c = multiple.mrc(  mrc , pmax=10, alpha = 2 , elbow.thresh = 0.5, n=500 )
p.hat = c$MDL
mrc.chpts = c$locs[[p.hat]][ c$affected[[p.hat]] ]
mrc.chpts
# }