QRMonFindAnomaliesByResiduals: Find anomalies by residuals.
In antononcube/QRMon-R: Quantile Regression Monad

Description Usage Arguments Details Value

Finds anomalies in the data by using a residuals of the fitted regression quantile and threshold or an outlier identifier.

QRMonFindAnomaliesByResiduals(
  qrObj,
  threshold = NULL,
  outlierIdentifier = NULL,
  relativeErrorsQ = FALSE,
  probability = NULL
)

`qrObj`	A QRMon object.
`threshold`	The threshold to be used to identify anomalies. If NULL the outlier identifier is used.
`outlierIdentifier`	Outlier identifier of list of numbers to be used to identify the anomalies. If NULL an internal implementation of the quartile outlier identifier is used. (For the top outliers.)
`relativeErrorsQ`	Should relative errors be used or not?
`probability`	Which regression quantile to use? If NULL the regression quantile is automatically selected, the one that is closest to 0.5.

The residuals outliers are picked with the formulas: c(xL,x0,xU) <- quantile( abs(residuals), c(1/4,1/2,3/4) ), abs(residuals) > x0+(xU-xL). (Errors and residuals are used as synonyms here.)

A QRMon object.

antononcube/QRMon-R documentation built on July 26, 2021, 1:07 p.m.