AnomalyDetectionTs: Anomaly Detection Using Seasonal Hybrid ESD Test

In Harminder858/anomaly_twitter_harminder: Anomaly Detection Using Seasonal Hybrid Extreme Studentized Deviate Test

Description

A technique for detecting anomalies in seasonal univariate time series where the input is a series of <timestamp, count> pairs.

Usage

AnomalyDetectionTs(x, max_anoms = 0.1, direction = "pos", alpha = 0.05,
  only_last = NULL, threshold = "None", e_value = FALSE,
  longterm = FALSE, piecewise_median_period_weeks = 2, verbose = FALSE,
  na.rm = FALSE)

ad_ts(x, max_anoms = 0.1, direction = "pos", alpha = 0.05,
  only_last = NULL, threshold = "None", e_value = FALSE,
  longterm = FALSE, piecewise_median_period_weeks = 2, verbose = FALSE,
  na.rm = FALSE)

Arguments

x	Time series as a two column data frame where the first column consists of the timestamps and the second column consists of the observations.
max_anoms	Maximum number of anomalies that S-H-ESD will detect as a percentage of the data.
direction	Directionality of the anomalies to be detected. One of: pos, neg, both.
alpha	The level of statistical significance with which to accept or reject anomalies.
only_last	Find and report anomalies only within the last day or hr in the time seriess. One of NULL, day, hr.
threshold	Only report positive going anoms above the threshold specified. One of: None, med_max, p95, p99.
e_value	Add an additional column to the anoms output containing the expected value.
longterm	Increase anom detection efficacy for time series that are greater than a month. See 'Details“ below.
piecewise_median_period_weeks	The piecewise median time window as described in Vallis, Hochenbaum, and Kejariwal (2014). Defaults to 2.
verbose	Enable debug messages
na.rm	Remove any NAs in timestamps.(default: FALSE)

Value

The returned value is a data frame containing timestamps, values, and optionally expected values.

References

• Vallis, O., Hochenbaum, J. and Kejariwal, A., (2014) "A Novel Technique for Long-Term Anomaly Detection in the Cloud", 6th USENIX, Philadelphia, PA. (https://www.usenix.org/system/files/conference/hotcloud14/hotcloud14-vallis.pdf)

• Rosner, B., (May 1983), "Percentage Points for a Generalized ESD Many-Outlier Procedure", Technometrics, 25(2), pp. 165-172. (https://www.jstor.org/stable/1268549)

See Also

ad_vec()

Examples

data(raw_data)

ad_ts(raw_data, max_anoms=0.02, direction='both')

# To detect only the anomalies on the last day, run the following:

ad_ts(raw_data, max_anoms=0.02, direction='both', only_last="day")

Harminder858/anomaly_twitter_harminder documentation built on March 29, 2020, 5:33 a.m.