RT Journal Article
SR Electronic
T1 A note on detecting statistical outliers in psychophysical data
JF bioRxiv
FD Cold Spring Harbor Laboratory
SP 074591
DO 10.1101/074591
A1 Pete R. Jones
YR 2016
UL http://biorxiv.org/content/early/2016/09/12/074591.abstract
AB This paper considers how best to identify statistical outliers in psychophysical datasets, where the underlying sampling distributions are unknown. Eight methods are described, and each is evaluated using Monte Carlo simulations of a typical psychophysical experiment. The best method is shown to be one based on a measure of absolute-deviation known as Sn. This method is shown to be more accurate than popular heuristics based on standard deviations from the mean, and more robust than non-parametric methods based on interquartile range. Matlab code for computing Sn is included.