cbar package

cbar: Contextual Bayesian Anomaly Detection in R

This function generates cbar object to detect contextual anomaly and to abstract analysis output.

cbar(.data, ref_period, mea_period, apply_standardized = T, interval = 0.95,
  ...)

Arguments

.data

data table with datetime, y, and predictors
ref_period

performance reference period
mea_period

performance measurement period
apply_standardized

whether it will standardized data or not
interval

credible interval. 0.95 by default.
...

params for bsts_model

Details

See the README on Github

For the input .data, note that you should use datetime for the first column name. Also, you should use numeric type for other columns.

Examples

library(cbar)

.data <- mtcars
rownames(.data) <- NULL
datetime <- seq(from = Sys.time(), length.out = nrow(.data), by = "mins")
.data <- cbind(datetime = datetime, .data)

ref_session <- 1:16
mea_session <- 17:nrow(.data)

obj <- cbar(.data, ref_session, mea_session)