Ticket #159 (closed defect: fixed)

Opened 2 months ago

Last modified 2 months ago

Kurtosis by Monte Carlo or distribution method

Reported by: willaume@phimeca.com Assigned to: lapointe@phimeca.com
Priority: trivial Milestone: openturns-0.13.0
Component: general Version: 0.12.1
Keywords: Cc:

Description

Depending on the method used, the value of the Kurtosis is either normalized or not.

Example in the attached file :

  • the method of the distribution class returns the value of the kurtosis : 5.4
  • using a numerical Sample (large enough) the value of the Kurtosis is 2.4

I know it is not a calculation error but according to me it would be better if the definition of the kurtosis is unique.

Attachments

test.py (311 bytes) - added by willaume@phimeca.com on 10/30/08 15:07:22.

Change History

10/30/08 15:07:22 changed by willaume@phimeca.com

  • attachment test.py added.

11/07/08 23:45:44 changed by regis.lebrun@eads.net

  • status changed from new to closed.
  • resolution set to fixed.
  • milestone changed from not defined to openturns-0.13.0.

This bug has already been fixed in the chaos branch (revision 998). The definition used for the NumericalSample is now the one used for the Distribution.

Régis