1

Original question is here.

I think the formula for population skewness from photo here is wrong.

I think correct formula is this.

Am i right?

[EDIT]

  1. Please check the example and the result in documentation with both formulas.
Human
  • 13
  • 3
    The formula in Excel's documentation is neither a formula for skewness nor is it what Excel actually does. It is just a typographical error in the documentation (this is obvious from the fact that there is a dashed-line square in the formula, which is a graphical placeholder for a term in Microsoft's equation editor and means nothing). – Chris Haug Jun 25 '21 at 19:04
  • @ChrisHaug That looks like an answer to me. – Dave Jun 25 '21 at 19:15

1 Answers1

2

First off, the formula in Excel's documentation is not really a valid formula for anything: there is a dashed-line square in there, which is a graphical placeholder for a term in Microsoft's equation editor and means nothing.

If you simply ignore it, you do not get a valid formula for skewness. The quickest way to see this is that the result is not scale-free (skewness is a "standardized" measure).

Fortunately, the formula in the documentation is not what Excel actually does, it is just a typographical error in the documentation. You can figure this out by taking a small dataset and computing it by hand with either formula and comparing to SKEW.P's result. I haven't tested it extensively, but it appears to be applying the other formula that you've shown, and it definitely does not use the one in the documentation.

Chris Haug
  • 5,785