| OutlierDetection {dr4pl} | R Documentation |
Detect outliers by the method of Motulsky and Brown (2006).
Description
Detect outliers by the method of Motulsky and Brown (2006).
Usage
OutlierDetection(residuals)
Arguments
residuals |
Vector of residuals from a robust fit. |
Details
This function detects outliers from a vector of residuals obtained from a robust fit. The method used here is the same with Motulsky and Brown (2006) except that the median absolute deviation is used instead of the sample quantile based estimator suggested in that paper. Based on the False Discovery Rate (FDR) a set of multiple outliers that have lower FDR's than a threshold are reported.
Value
Vector of indices of outliers in the input vector of residuals
Author(s)
Hyowon An
References
Motulsky HJ, Brown RE (2006). “Detecting outliers when fitting data with nonlinear regression - a new method based on robust nonlinear regression and the false discovery rate.” BMC Bioinformatics, 7, 123.