Exact mathematical formulas for Weighted SKEW() and Weighted KURTOSIS() chart functions (weighted calculations)

[martinmich]

Hello experts.

There could be different mathematical formulas for SKEW() and KURTOSIS() chart functions, e.g. just 3 different definitions are mentioned 1.3.5.11. Measures of Skewness and Kurtosis (nist.gov)

There is also excellent explanation from Qlik Community what exact mathematical formulas are used for the non-weighted calculations. And video at Data Voyagers answering the mechanics behind the mathematical formulas.

However it's not clear if it is still possible to use WEIGHTS within the formulas for the data? What I mean by this is that each row of data would have some numerical weight value which is then considered for the final calculation of Skew or Kurtosis (in a similar logic like weighted average or weighted mean).

Unfortunately, For Kurtosis and Skew functions official documentation does not mention weighted options:
https://help.qlik.com/en-US/sense/May2024/Subsystems/Hub/Content/Sense_Hub/ChartFunctions/Statistica...

https://help.qlik.com/en-US/sense/May2024/Subsystems/Hub/Content/Sense_Hub/ChartFunctions/Statistica...

Can Qlik developers check and provide some clarity on this? 

Many thanks ahead. Regards, Martin 

[igoralcantara]

Unless you write a formula in your load script to add the weight, it can't be done. I mean, Skew and Kurtosis functions do not have that parameter.

[martinmich]

I though the same but this could probably increase the time for calculations and it needs to be done on row by row basis. would you have any example how to do the calculation and what formula to use in the chart? it seems we will need to do on-the-fly calculations.