SNSeg_Uni: Self-normalization (SN) based change point estimates for...

In SNSeg: Self-Normalization(SN) Based Change-Point Estimation for Time Series

Self-normalization (SN) based change point estimates for univariate time series

Description

The function SNSeg_Uni is a SN change point estimation procedure for a univariate time series based on the change in a single or multiple parameters . It also detect changes in correlation between two univariate time series.

Usage

SNSeg_Uni(
  ts,
  paras_to_test,
  confidence = 0.9,
  grid_size_scale = 0.05,
  grid_size = NULL,
  plot_SN = TRUE,
  est_cp_loc = TRUE
)

Arguments

ts	A univariate time series expressed as a numeric vector. when the argument paras_to_test is specified as "bivcor", the correlation between bivariate time series, the input ts must be an n by 2 matrix
paras_to_test	The parameters that SN algorithm aim to examine, which are presented as a string, a number, or a combination of both. Available choices of paras_to_test include "mean", "variance", "acf", "bivcor" and a numeric value of quantile between 0 and 1. In the scenario where the input ts is a univariate time series, users are allowed to enter a combination of parameters for paras_to_test except "bivcor". Users can also set up their own function as the input of "paras_to_test". If so, the user-fined function should use the univariate time series as the input and return a numeric value as the output. Please see the help vignette for more details by running vignette("SNSeg", package = "SNSeg").
confidence	Confidence level of SN tests as a numeric value. Available choices of confidence levels contain 0.9, 0.95, 0.99, 0.995 and 0.999. The default is set to 0.9.
grid_size_scale	A numeric value of the trimming parameter and only in use if grid_size = NULL. Users are allowed to choose any grid_size_scale between 0.05 and 0.5. A warning will be given if it is outside the range.
grid_size	Local window size h to compute the critical value for SN test. Since grid_size = n*grid_size_scale, where n is the length of time series, this function will compute the grid_size_scale by dividing n from grid_size when it is not NULL.
plot_SN	Boolean value to plot the time series or not. The default setting is FALSE.
est_cp_loc	Boolean value to plot a red solid vertical line for estimated change-point locations if est_cp_loc = TRUE

Value

SNSeg_Uni returns an S3 object of class "SNSeg_Uni" including the time series, the type of parameter to be tested, the local window size to cover a change point, the estimated change-point locations, the confidence level and the critical value of the SN test. It also generates a time series segmentation plot when plot_SN = TRUE.

ts

A numeric vector or two-dimensional matrix of the input time series.

paras_to_test

A character, numeric value, a function or vector of the parameter(s) used for the SN test. If it is a function defined by the user, please refer to the section "test in a general functional" in the help vignette for more details on how to write the function correctly.

grid_size

A numeric value of the window size.

SN_sweep_result

A list of matrices where each matrix consists of four columns: (1) SN-based test statistic for each change-point location (2) Change-point location (3) Lower bound of the local window and (4) Upper bound of the local window.

est_cp

A vector containing the locations of the estimated change-points.

confidence

Confidence level of SN test as a numeric value.

critical_value

Critical value of the SN-based test statistic.

Users can apply the functions summary.SN to compute the parameter estimate of each segment separated by the detected change-points. An additional function plot.SN can be used to plot the time series with estimated change-points. Users can set the option plot_SN = TRUE or use the function plot.SN to plot the time series.

It deserves to note that some change-points could be missing due to the constraint on grid_size_scale or related grid_size that grid_size_scale has a minimum value of 0.05. Therefore, SNCP claims no change-points within the first ngrid_size_scale or the last ngrid_size_scale time points. This is a limitation of the function SNSeg_Uni.

For more examples of SNSeg_Uni see the help vignette: vignette("SNSeg", package = "SNSeg")

Examples

# code to simulate a univariate time series
set.seed(7)
ts <- MAR_Variance(2, "V1")
ts <- ts[,2]
# test the change in a single parameter (variance)
# grid_size defined
result <- SNSeg_Uni(ts, paras_to_test = "variance", confidence = 0.9,
                    grid_size_scale = 0.05, grid_size = 67,
                    plot_SN = TRUE, est_cp_loc = TRUE)
# estimated change-point locations
result$est_cp
# For more examples of change in a single or multiple parameters, please run
# the command: vignette("SNSeg", package = "SNSeg")

SNSeg documentation built on June 22, 2024, 10:50 a.m.