fastcpd (version 0.16.0)

fastcpd_meanvariance: Find change points efficiently in mean variance change models

Description

fastcpd_meanvariance(), fastcpd.meanvariance(), fastcpd_mv(), fastcpd.mv() are wrapper functions of fastcpd() to find the meanvariance change. The function is similar to fastcpd() except that the data is by default a matrix or data frame or a vector with each row / element as an observation and thus a formula is not required here.

Usage

fastcpd_meanvariance(data, ...)
fastcpd.meanvariance(data, ...)
fastcpd_mv(data, ...)
fastcpd.mv(data, ...)

Value

A fastcpd object.

Arguments

data

A matrix, a data frame or a vector.

...

Other arguments passed to fastcpd(), for example, segment_count.

See Also

fastcpd()

Examples

set.seed(1)
p <- 3
data <- if (requireNamespace("mvtnorm", quietly = TRUE)) {
  rbind(
    mvtnorm::rmvnorm(2e+5, mean = rep(0, p), sigma = diag(1, p)),
    mvtnorm::rmvnorm(1e+5, mean = rep(50, p), sigma = diag(1, p)),
    mvtnorm::rmvnorm(2e+5, mean = rep(0, p), sigma = diag(100, p)),
    mvtnorm::rmvnorm(2e+5, mean = rep(0, p), sigma = diag(1, p)),
    mvtnorm::rmvnorm(1e+5, mean = rep(50, p), sigma = diag(1, p)),
    mvtnorm::rmvnorm(2e+5, mean = rep(50, p), sigma = diag(100, p))
  )
} else {
  rbind(
    matrix(rnorm(p * 2e+5, mean = 0, sd = 1), ncol = p),
    matrix(rnorm(p * 1e+5, mean = 50, sd = 1), ncol = p),
    matrix(rnorm(p * 2e+5, mean = 0, sd = 10), ncol = p),
    matrix(rnorm(p * 2e+5, mean = 0, sd = 1), ncol = p),
    matrix(rnorm(p * 1e+5, mean = 50, sd = 1), ncol = p),
    matrix(rnorm(p * 2e+5, mean = 50, sd = 10), ncol = p)
  )
}
(result_time <- system.time(result <- fastcpd.mv(data)))
summary(result)
result@thetas[seq_len(p), ]
lapply(result@thetas[seq_len(p^2) + p, ], function(thetas) matrix(thetas, p))