deanomalize: Replace Anomalies with Expected Values Using the 'anomalize'...

In jesse-smith/covidModel: COVID-19 Modeling Tools for Shelby County Health Department

Description

deanomalize() wraps the workflow from the anomalize package into a single function. This includes decomposition into seasonal, trend, and remainder components using robust STL; identification of anomalies in the remainders using an iterative Generalize Extreme Studentized Deviate Test; and replacment of the identified anomalies with the seasonal + trend expected value.

Usage

deanomalize(
  .data,
  .col,
  period = NULL,
  trend = NULL,
  cutoff = 0.05,
  quiet = FALSE,
  plot = FALSE,
  ...
)

Arguments

.data	A data frame containg a time-based column and a column of observations. The time-based column will be identified automatically. Can include other columns, but these will be dropped.
.col	The column containing the observations at each time point
period	The length of time to use in seasonal decomposition; can be a time-based definition (e.g. "1 week") or an integer number of days. If NULL or "auto", period is set automatically using the tunable heuristics in the timetk package.
trend	The length of time to use in trend decomposition; can be a time-based definition (e.g. "1 month") or an integer number of days. If NULL or "auto", trend is set automatically using the tunable heuristics in the timetk package.
cutoff	The cutoff value for anomaly detection; controls both the maximum percentage of data points that may be considered anomalies, as well as the critical value for the Generalized Extreme Studentized Deviate test used to detect the anomalies. Can be interpreted as the desired maximum probability that an individual data point is labeled an anomaly.
quiet	Should messages and warnings be suppressed?
plot	Should anomalies be plotted for visual inspection?
...	Additional arguments to pass to time_decompose(method = "stl")

Value

A tibble with a column for the time index and columns observed, season, trend, remainder, remainder_l1, remainder_l2, anomaly, and observed_cleaned

jesse-smith/covidModel documentation built on Feb. 21, 2022, 3:23 p.m.