02-16-2019 13:33 PM - last edited 02-16-2019 13:34 PM
Simply stated, Kurtosis is a measure of the degree to which a given distribution is more or less "peaked" relative to the normal distribution. This is an important measure that can have implications on what sorts of statistical methods are appropriate to use on the sample distribution.
You can read more about Kurtosis at these links:
The presented Quick Measure has three measures, Sample Kurtosis, Excess Kurtosis and Kurtosis Type:
Sample Kurtosis = VAR __mean = AVERAGEX(ALL('Data'),[Values]) VAR __stddev = STDEVX.P(ALL('Data'),[Values]) VAR __n = COUNTROWS(ALL('Data')) VAR __table = ADDCOLUMNS('Data',"__skew",POWER(([Values]-__mean),4)) VAR __sum = SUMX(__table,[__skew]) RETURN DIVIDE(__sum,POWER(__stddev,4),0) * DIVIDE(1,__n,0)
Excess Kurtosis = [Sample Kurtosis] - 3
Kurtosis Type = SWITCH(TRUE(), [Excess Kurtosis]>0,"Leptokurtic", [Excess Kurtosis]<0,"Platykurtic", "Mesokurtic" )
Proud to be a Datanaut!