anomaly_thresh

A helper function that wraps around <code>quantile</code> to apply a threshold to anomaly scores.

Apply the spectral residual algorithm to data, such as a
time series, to detect anomalies. Anomaly scores can be used to determine
outliers based upon a threshold or fed into more sophisticated prediction
models. Methods are based upon "Time-Series Anomaly Detection Service at
Microsoft", Ren, H., Xu, B., Wang, Y., et al., (2019)
<doi:10.48550/arXiv.1906.03821>.

Allen OBrien

spectralAnomaly

Detect Anomalies Using the Spectral Residual Algorithm

anomaly_thresh function

<dl><dt>x</dt>
<dd>Numeric vector of anomaly scores (e.g. created by <code>anomaly_score</code>).</dd>
<dt>threshold</dt>
<dd>Numeric value to determine the threshold to flag outliers among the score.</dd>
<dt>...</dt>
<dd>Additional parameters passed to <code>quantile</code>.</dd></dl>

Arguments

Apply threshold to anomaly score — anomaly_thresh

<dl>

<dt>x</dt>
<dd>Numeric vector of anomaly scores (e.g. created by <code>anomaly_score</code>).</dd>


<dt>threshold</dt>
<dd>Numeric value to determine the threshold to flag outliers among the score.</dd>


<dt>...</dt>
<dd>Additional parameters passed to <code>quantile</code>.</dd>

</dl>

anomaly_thresh: Apply threshold to anomaly score

Description

Usage

Value

Arguments

Examples