detect_cp_uni: Detect Change Points on an univariate time series.

Description

Detect Change Points on an univariate time series.

Usage

detect_cp_uni(
  data,
  n_iterations,
  q,
  a = 1,
  b = 1,
  c = 0.1,
  prior_var_phi = 0.1,
  prior_delta_c = 1,
  prior_delta_d = 1,
  print_progress = TRUE,
  user_seed = 1234L
)

Value

Function detect_cp_uni returns a list containing the following components:

$orders a matrix where each row corresponds to the output order of the corresponding iteration.
time computational time in seconds.
$sigma_MCMC traceplot for $\sigma$.
$sigma_MCMC_01 a $0/1$ vector, the $n$-th element is equal to $1$ if the proposed $\sigma$ was accepted, $0$ otherwise.
$delta_MCMC traceplot for $\delta$.

Arguments

data: vector of observations.
n_iterations: number of MCMC iteration.
q: probability of performing a split at each iterations.
a, b, c: parameters of the Normal-Gamma prior for $\mu$ and $\lambda$.
prior_var_phi: parameters for the correlation coefficient in the likelihood.
prior_delta_c, prior_delta_d: parameters of the shifted Gamma prior for $\delta$.
print_progress: If TRUE (default) print the progress bar.
user_seed: seed for random distribution generation.

Examples

Run this code


data_vec <- as.numeric(c(rnorm(50,0,0.1), rnorm(50,1,0.25)))

out <- detect_cp_uni(data = data_vec,
                            n_iterations = 2500,
                            q = 0.25)

Run the code above in your browser using DataLab