An outlier index (OI) is defined for DatLab analysis, derived from Pearson’s coefficient of skewness, but more specific in targeting outliers in data series recorded with the O2k. At the limit of a zero value, Y = ABS(Average + Median)/2, the OI equals Pearson’s coefficient of skewness #2 (without the multiplication factor of 3). At high Y with small standard deviation (SD), the outlier index is effectively the difference between the Average and the Median normalized for the absolute value, (Average-Median)/Y. The definition of the outlier index is,
- OI = (Average-Median)/(Y + SD)
- OI = (Average-Median)/[ABS(Average+Median)/2 + SD]
- Pearson’s coefficient of skewness #2 = 3 x (Average-Median)/SD
The threshold of the absolute value of the OI is set at 0.05. If ABS(OI)>0.05 calculated for the data points within a defined Mark, the Mark window indicates the likely occurrence of outliers in the data sequence. The threshold can be set to a lab-specific or session-specific value different from the default value.
Communicated by Gnaiger E 2016-10-03; updated 2016-10-22.