Pipe operator

Description

See magrittr::%>% for details.

Usage

lhs %>% rhs

Generic functions and methods: SC

Description

Generic functions and methods: SC

Usage

SC(x)

SC(x) <- value

## S4 method for signature 'cpss'
SC(x)

## S4 replacement method for signature 'cpss'
SC(x) <- value

Arguments

Generic functions and methods: SC_vals

Description

Generic functions and methods: SC_vals

Usage

SC_vals(x)

SC_vals(x) <- value

## S4 method for signature 'cpss'
SC_vals(x)

## S4 replacement method for signature 'cpss'
SC_vals(x) <- value

Arguments

Generic functions and methods: S_vals

Description

Generic functions and methods: S_vals

Usage

S_vals(x)

S_vals(x) <- value

## S4 method for signature 'cpss'
S_vals(x)

## S4 replacement method for signature 'cpss'
S_vals(x) <- value

Arguments

Generic functions and methods: algo

Description

Generic functions and methods: algo

Usage

algo(x)

algo(x) <- value

## S4 method for signature 'cpss'
algo(x)

## S4 replacement method for signature 'cpss'
algo(x) <- value

Arguments

Generic functions and methods: algo_param_dim

Description

Generic functions and methods: algo_param_dim

Usage

algo_param_dim(x)

algo_param_dim(x) <- value

## S4 method for signature 'cpss'
algo_param_dim(x)

## S4 replacement method for signature 'cpss'
algo_param_dim(x) <- value

Arguments

coef method

Description

coef method

Usage

## S4 method for signature 'cpss'
coef(object)

Arguments

Generic functions and methods: cps

Description

Generic functions and methods: cps

Usage

cps(x)

cps(x) <- value

## S4 method for signature 'cpss'
cps(x)

## S4 replacement method for signature 'cpss'
cps(x) <- value

Arguments

cpss: Change-Point Detection by Sample-Splitting Methods

Description

Implements multiple change searching algorithms for a variety of frequently considered parametric change-point models. In particular, it integrates a criterion proposed by Zou, Wang and Li (2020) doi:10.1214/19-AOS1814 to select the number of change-points in a data-driven fashion. Moreover, it also provides interfaces for user-customized change-point models with one's own cost function and parameter estimation routine.

Getting started

Easy to get started with the cpss.* set of functions by accessing their documentation pages
library(cpss)
?cpss.mean
?cpss.var
?cpss.meanvar
?cpss.glm
?cpss.lm
?cpss.em
?cpss.custom

cpss: an S4 class which collects data and information required for further change-point analyses and summaries

Description

cpss: an S4 class which collects data and information required for further change-point analyses and summaries

Slots

dat

ANY.

mdl

character.

algo

character.

algo_param_dim

numeric.

SC

character.

ncps

integer.

pelt_pen

numeric.

cps

numeric.

params

list.

S_vals

numeric.

SC_vals

matrix.

call

list.

update_inputs

list.

Detecting changes in uers-customized models

Description

Detecting changes in uers-customized models

Usage

cpss.custom(
  dataset,
  n,
  g_subdat,
  g_param,
  g_cost,
  algorithm = "BS",
  dist_min = floor(log(n)),
  ncps_max = ceiling(n^0.4),
  pelt_pen_val = NULL,
  pelt_K = 0,
  wbs_nintervals = 500,
  criterion = "CV",
  times = 2,
  model = NULL,
  g_smry = NULL,
  easy_cost = NULL,
  param.opt = NULL
)

Arguments

Value

cpss.custom returns an object of an S4 class, called "cpss", which collects data and information required for further change-point analyses and summaries.

dat

data set

mdl

considered change-point model

algo

change-point searching algorithm

algo_param_dim

user-specified upper bound of the number of true change-points if algorithm = "SN"/"BS"/"WBS", or user-specified candidate values of the penalty only if algorithm = "PELT"

SC

model selection criterion

ncps

estimated number of change-points

pelt_pen

selected value of the penalty only if algorithm = "PELT"

cps

a vector of estimated locations of change-points

params

a list object, each member is a list containing estimated parameters in the associated data segment

S_vals

a numeric vector of candidate model dimensions in terms of a sequence of numbers of change-points or values of the penalty

SC_vals

a numeric matrix, each column records the values of the criterion based on the validation data split under the corresponding model dimension (S_vals), and each row represents a splitting at each time

References

Killick, R., Fearnhead, P., and Eckley, I. A. (2012). Optimal Detection of Changepoints With a Linear Computational Cost. Journal of the American Statistical Association, 107(500): 1590–1598.

Fryzlewicz, P. (2014). Wild binary segmentation for multiple change-point detection. The Annals of Statistics, 42(6): 2243–2281.

Examples

library("cpss")
g_subdat_l1 <- function(dat, indices) {
  dat[indices]
}
g_param_l1 <- function(dat, param.opt = NULL) {
  return(median(dat))
}
g_cost_l1 <- function(dat, param) {
  return(sum(abs(dat - param)))
}
res <- cpss.custom(
  dataset = well, n = length(well),
  g_subdat = g_subdat_l1, g_param = g_param_l1, g_cost = g_cost_l1,
  ncps_max = 11
)
summary(res)
plot(well)
abline(v = res@cps, col = "red")

Detecting changes in exponential family

Description

Detecting changes in exponential family

Usage

cpss.em(
  dataset,
  family,
  size = NULL,
  algorithm = "BS",
  dist_min = floor(log(n)),
  ncps_max = ceiling(n^0.4),
  pelt_pen_val = NULL,
  pelt_K = 0,
  wbs_nintervals = 500,
  criterion = "CV",
  times = 2
)

Arguments

Value

cpss.em returns an object of an S4 class, called "cpss", which collects data and information required for further change-point analyses and summaries. See cpss.custom.

References

Killick, R., Fearnhead, P., and Eckley, I. A. (2012). Optimal Detection of Changepoints With a Linear Computational Cost. Journal of the American Statistical Association, 107(500):1590–1598.

Fryzlewicz, P. (2014). Wild binary segmentation for multiple change-point detection. The Annals of Statistics, 42(6): 2243–2281.

See Also

cpss.meanvar cpss.mean cpss.var

Examples

library("cpss")
set.seed(666)
n <- 1000
tau <- c(100, 300, 700, 900)
tau_ext <- c(0, tau, n)
theta <- c(1, 0.2, 1, 0.2, 1)
seg_len <- diff(c(0, tau, n))
y <- unlist(lapply(seq(1, length(tau) + 1), function(k) {
  rexp(seg_len[k], theta[k])
}))
res <- cpss.em(
  y, family = "exp", algorithm = "WBS", ncps_max = 10,
  criterion = "MS", times = 10
)
cps(res)
# [1] 100 299 705 901

Detecting changes in GLMs

Description

Detecting changes in GLMs

Usage

cpss.glm(
  formula,
  family,
  data = NULL,
  algorithm = "BS",
  dist_min = floor(log(n)),
  ncps_max = ceiling(n^0.4),
  pelt_pen_val = NULL,
  pelt_K = 0,
  wbs_nintervals = 500,
  criterion = "CV",
  times = 2
)

Arguments

Value

cpss.glm returns an object of an S4 class, called "cpss", which collects data and information required for further change-point analyses and summaries. See cpss.custom.

References

Killick, R., Fearnhead, P., and Eckley, I. A. (2012). Optimal Detection of Changepoints With a Linear Computational Cost. Journal of the American Statistical Association, 107(500):1590–1598.

Fryzlewicz, P. (2014). Wild binary segmentation for multiple change-point detection. The Annals of Statistics, 42(6): 2243–2281.

See Also

cpss.lm

Examples

library("cpss")
set.seed(666)
n <- 200
size <- rpois(n, 20 - 1) + 1
tau <- c(75, 100, 175)
tau_ext <- c(0, tau, n)
be <- list(c(0, 0.5), c(0, -0.5), c(0.5, -0.5), c(-0.5, -0.5))
seg_len <- diff(c(0, tau, n))
x <- rnorm(n)
eta <- lapply(seq(1, length(tau) + 1), function(k) {
  be[[k]][1] + be[[k]][2] * x[(tau_ext[k] + 1):tau_ext[k + 1]]
})
eta <- do.call(c, eta)
p <- 1 / (1 + exp(-eta))
y <- rbinom(n, size = size, prob = p)

pelt_pen_val <- (log(n))^seq(0.5, 2, by = 0.1)
res <- cpss.glm(
  formula = cbind(y, size - y) ~ x, family = binomial(),
  algorithm = "PELT", pelt_pen_val = pelt_pen_val, ncps_max = 10
)
summary(res)
# 75  105  175
coef(res)
# [1,] 0.02540872  0.08389551  0.5284425 -0.4980768
# [2,] 0.57222684 -0.45430385 -0.5203319 -0.4581678

Detecting changes in linear models

Description

Detecting changes in linear models

Usage

cpss.lm(
  formula,
  data = NULL,
  algorithm = "BS",
  dist_min = floor(log(n)),
  ncps_max = ceiling(n^0.4),
  pelt_pen_val = NULL,
  pelt_K = 0,
  wbs_nintervals = 500,
  criterion = "CV",
  times = 2
)

Arguments

Value

cpss.lm returns an object of an S4 class, called "cpss", which collects data and information required for further change-point analyses and summaries. See cpss.custom.

References

Killick, R., Fearnhead, P., and Eckley, I. A. (2012). Optimal Detection of Changepoints With a Linear Computational Cost. Journal of the American Statistical Association, 107(500):1590–1598.

Fryzlewicz, P. (2014). Wild binary segmentation for multiple change-point detection. The Annals of Statistics, 42(6): 2243–2281.

See Also

cpss.glm

Examples

library("cpss")
set.seed(666)
n <- 400
tau <- c(80, 200, 300)
tau_ext <- c(0, tau, n)
be <- list(c(0, 1), c(1, 0.5), c(0, 1), c(-1, 0.5))
seg_len <- diff(c(0, tau, n))
x <- rnorm(n)
mu <- lapply(seq(1, length(tau) + 1), function(k) {
  be[[k]][1] + be[[k]][2] * x[(tau_ext[k] + 1):tau_ext[k + 1]]
})
mu <- do.call(c, mu)
sig <- unlist(lapply(seq(1, length(tau) + 1), function(k) {
  rep(be[[k]][2], seg_len[k])
}))
y <- rnorm(n, mu, sig)
res <- cpss.lm(
  formula = y ~ x,
  algorithm = "BS", ncps_max = 10
)
summary(res)
# 80  202  291
coef(res)
# $coef
#             [,1]      [,2]        [,3]       [,4]
# [1,] -0.00188792 1.0457718 -0.03963209 -0.9444813
# [2,]  0.91061557 0.6291965  1.20694409  0.4410036
#
# $sigma
# [1] 0.8732233 0.4753216 0.9566516 0.4782329

Detecting changes in mean

Description

Detecting changes in mean

Usage

cpss.mean(
  dataset,
  algorithm = "BS",
  dist_min = floor(log(n)),
  ncps_max = ceiling(n^0.4),
  pelt_pen_val = NULL,
  pelt_K = 0,
  wbs_nintervals = 500,
  criterion = "CV",
  times = 2,
  Sigma = NULL
)

Arguments

Value

cpss.mean returns an object of an S4 class, called "cpss", which collects data and information required for further change-point analyses and summaries. See cpss.custom.

References

Killick, R., Fearnhead, P., and Eckley, I. A. (2012). Optimal Detection of Changepoints With a Linear Computational Cost. Journal of the American Statistical Association, 107(500): 1590–1598. Fryzlewicz, P. (2014). Wild binary segmentation for multiple change-point detection. The Annals of Statistics, 42(6): 2243–2281.

See Also

cpss.meanvar cpss.var

Examples

library("cpss")
set.seed(666)
n <- 2048
tau <- c(205, 267, 308, 472, 512, 820, 902, 1332, 1557, 1598, 1659)
seg_len <- diff(c(0, tau, n))
mu <- rep(c(0, 14.64, -3.66, 7.32, -7.32, 10.98, -4.39, 3.29, 19.03, 7.68, 15.37, 0), seg_len)
ep <- 7 * rnorm(n)
y <- mu + ep

res <- cpss.mean(y, algorithm = "SN", ncps_max = 20)
summary(res)
# 205  267  307  471  512  820  897  1332  1557  1601  1659
plot(res, type = "scatter")
plot(res, type = "path")
out <- update(res, dim_update = 12)
out@cps
# 205  267  307  471  512  820  897 1332 1557 1601 1659 1769
# coef(out)

Detecting changes in mean and (co)variance

Description

Detecting changes in mean and (co)variance

Usage

cpss.meanvar(
  dataset,
  algorithm = "BS",
  dist_min = floor(log(n)),
  ncps_max = ceiling(n^0.4),
  pelt_pen_val = NULL,
  pelt_K = 0,
  wbs_nintervals = 500,
  criterion = "CV",
  times = 2
)

Arguments

Value

cpss.meanvar returns an object of an S4 class, called "cpss", which collects data and information required for further change-point analyses and summaries. See cpss.custom.

References

Killick, R., Fearnhead, P., and Eckley, I. A. (2012). Optimal Detection of Changepoints With a Linear Computational Cost. Journal of the American Statistical Association, 107(500):1590–1598. Fryzlewicz, P. (2014). Wild binary segmentation for multiple change-point detection. The Annals of Statistics, 42(6): 2243–2281.

See Also

cpss.mean cpss.var

Examples

library("cpss")
if (!requireNamespace("MASS", quietly = TRUE)) {
  stop("Please install the package \"MASS\".")
}
set.seed(666)
n <- 1000
tau <- c(200, 400, 600, 800)
mu <- list(rep(0, 2), rep(1, 2), rep(1, 2), rep(0, 2), rep(0, 2))
Sigma <- list(diag(2), diag(2), matrix(c(1,-1,-1, 4), 2), matrix(c(1, 0.5, 0.5, 1), 2), diag(2))
seg_len <- diff(c(0, tau, n))
y <- lapply(seq(1, length(tau) + 1), function(k) {
  MASS::mvrnorm(n = seg_len[k], mu = mu[[k]], Sigma = Sigma[[k]])
})
y <- do.call(rbind, y)
res <- cpss.meanvar(y, algorithm = "BS", dist_min = 20)
cps(res)
# [1] 211 402 598 804
plot(res, type = "coef")

Detecting changes in (co)variance

Description

Detecting changes in (co)variance

Usage

cpss.var(
  dataset,
  algorithm = "BS",
  dist_min = floor(log(n)),
  ncps_max = ceiling(n^0.4),
  pelt_pen_val = NULL,
  pelt_K = 0,
  wbs_nintervals = 500,
  criterion = "CV",
  times = 2,
  mu = NULL
)

Arguments

Value

cpss.var returns an object of an S4 class, called "cpss", which collects data and information required for further change-point analyses and summaries. See cpss.custom.

References

Killick, R., Fearnhead, P., and Eckley, I. A. (2012). Optimal Detection of Changepoints With a Linear Computational Cost. Journal of the American Statistical Association, 107(500): 1590–1598. Fryzlewicz, P. (2014). Wild binary segmentation for multiple change-point detection. The Annals of Statistics, 42(6): 2243–2281.

See Also

cpss.meanvar cpss.mean

Examples

library("cpss")
if (!requireNamespace("MASS", quietly = TRUE)) {
  stop("Please install the package \"MASS\".")
}
set.seed(666)
n <- 1000
tau <- c(200, 500, 750)
mu <- list(rep(0, 2), rep(0, 2), rep(0, 2), rep(0, 2))
Sigma <- list(diag(2), matrix(c(1, 0, 0, 4), 2), matrix(c(1, -0.5, -0.5, 4), 2), diag(2))
seg_len <- diff(c(0, tau, n))
y <- lapply(seq(1, length(tau) + 1), function(k) {
  MASS::mvrnorm(n = seg_len[k], mu = mu[[k]], Sigma = Sigma[[k]])
})
y <- do.call(rbind, y)
res <- cpss.var(y, algorithm = "BS", dist_min = 20)
cps(res)
# [1] 215 515 751

Generic functions and methods: dat

Description

Generic functions and methods: dat

Usage

dat(x)

dat(x) <- value

## S4 method for signature 'cpss'
dat(x)

## S4 replacement method for signature 'cpss'
dat(x) <- value

Arguments

Generic functions and methods: mdl

Description

Generic functions and methods: mdl

Usage

mdl(x)

mdl(x) <- value

## S4 method for signature 'cpss'
mdl(x)

## S4 replacement method for signature 'cpss'
mdl(x) <- value

Arguments

Generic functions and methods: ncps

Description

Generic functions and methods: ncps

Usage

ncps(x)

ncps(x) <- value

## S4 method for signature 'cpss'
ncps(x)

## S4 replacement method for signature 'cpss'
ncps(x) <- value

Arguments

Generic functions and methods: params

Description

Generic functions and methods: params

Usage

params(x)

params(x) <- value

## S4 method for signature 'cpss'
params(x)

## S4 replacement method for signature 'cpss'
params(x) <- value

Arguments

Generic functions and methods: pelt_pen

Description

Generic functions and methods: pelt_pen

Usage

pelt_pen(x)

pelt_pen(x) <- value

## S4 method for signature 'cpss'
pelt_pen(x)

## S4 replacement method for signature 'cpss'
pelt_pen(x) <- value

Arguments

plot method

Description

plot method

Usage

## S4 method for signature 'cpss'
plot(obj, type, x = c(), y = c(), ...)

Arguments

summary method

Description

summary method

Usage

## S4 method for signature 'cpss'
summary(object)

Arguments

update method

Description

update method

Usage

## S4 method for signature 'cpss'
update(object, dim_update)

Arguments

Generic functions and methods: update_inputs

Description

Generic functions and methods: update_inputs

Usage

update_inputs(x)

update_inputs(x) <- value

## S4 method for signature 'cpss'
update_inputs(x)

## S4 replacement method for signature 'cpss'
update_inputs(x) <- value

Arguments

Well-log data

Description

Measurements of the nuclear magnetic response of underground rocks.

Usage

well

Format

A vector of 4,050 measurements:

well

Measurements.

Source

doi:10.1111/1467-9868.00421