hanr_garch: Anomaly detector using GARCH

In harbinger: A Unified Time Series Event Detection Framework

Anomaly detector using GARCH

Description

Fits a GARCH model to capture conditional heteroskedasticity and flags observations with large standardized residuals as anomalies. Wraps rugarch.

Usage

hanr_garch()

Details

A sGARCH(1,1) with ARMA(1,1) mean is estimated. Standardized residuals are summarized and thresholded via harutils().

Value

hanr_garch object.

References

• Engle RF (1982). Autoregressive Conditional Heteroscedasticity with Estimates of the Variance of United Kingdom Inflation. Econometrica, 50(4):987–1007.

• Bollerslev T (1986). Generalized Autoregressive Conditional Heteroskedasticity. Journal of Econometrics, 31(3):307–327.

Examples

library(daltoolbox)

# Load anomaly example data
data(examples_anomalies)

# Use a simple example
dataset <- examples_anomalies$simple
head(dataset)

# Configure GARCH anomaly detector
model <- hanr_garch()

# Fit the model
model <- fit(model, dataset$serie)

# Run detection
detection <- detect(model, dataset$serie)

# Show detected anomalies
print(detection[(detection$event),])

harbinger documentation built on Feb. 11, 2026, 9:06 a.m.