BFF_gaussian_predictive: BFF Change Point Detector using Predictive Posterior P-Values...

In elizabethriddle/BFF: Sequential Adaptive Parameter Estimation and Change Point Detection

BFF Change Point Detector using Predictive Posterior P-Values for Gaussian Data

Description

Detects change points within data stream sequentially using a Bayesian adaptive estimation procedure and predictive posterior p-values.

Usage

BFF_gaussian_predictive(
  data,
  threshold_val = c(0.05),
  burnin = 200,
  grace_period = 20,
  alpha = 39,
  beta = 1.8,
  sw = 2000,
  collect_p_val = TRUE
)

Arguments

data	data stream to perform change point detection on
threshold_val	vector of thresholds to use to assess whether new data point is a change point
burnin	the intial period to use to build the model on which no change points are detected
grace_period	period after a change in which changes are not detected
alpha	alpha parameter for lambda beta prior
beta	beta parameter for lambda beta prior
sw	sliding window size for p-value calibration
collect_p_val	whether or not to store p-values, boolean

Value

return the change points detected by the algorithm for each of the dection procedures, for the different thresholds. If collecting p-values then results will contain these.

elizabethriddle/BFF documentation built on April 4, 2022, 11:51 a.m.