[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[EquisMetaStock Group] Re: Skewness or biasedness and kurtosis..MG Ferreira and



PureBytes Links

Trading Reference Links


Thanks, please note, I have no problem at all with your
version, apart from that it is not what a textbook will
give you as kurtosis.  This difference is probably an
implementation trick ;-) I'll look at your formula a bit,
to try and understand how you measured this....

Just a further comment, for what it is worth.  Yes,
kurtosis measures the peakedness, but the really
interesting part of the distribution will be the tails.
If the distro is more peaked (leptokurtic) then the
tails are more fat - which is usually the case with
financial market data.  So kurtosis, although strictly
speaking a measure of peakedness, is really used to
understand behaviour in the tails, as is skewness.

Regards
MG

--- In equismetastock@xxxxxxxxxxxxxxx, Dusant <cooldush@xxxx> wrote:
> The "exact" Kurtosis (as you call it), creates "windows" of spikes
based on the look back periods. The formula presented by me is a
simpler version, which avoids those spikes in the Kurtosis.
>  
> As you may be aware, Kurtosis measures the "peak" of a distribution,
using the mean and the standard deviation. Again it depends on what
method of mean you are taking from the distribution. Whether you are
taking the mean of a single day, the Moving Average of the mean over a
certain period, or the mathematical mean over a certain period.
>  
> Its very subjective, and again user specific. The kurtosis is
supposed to give a bell curve, and analysed.
>  
> This is my very humble submission, of what little I know of
kurtosis, and I admit I my be wrong.
>  
> Dusant
>   ----- Original Message ----- 
>   From:   MG   Ferreira 
>   To: equismetastock@xxxx   
>   Sent: Monday, March 07, 2005 11:44   AM
>   Subject: [EquisMetaStock Group] Re:   Skewness or biasedness and
kurtosis..MG Ferreira and Dusant
>   
> 
> 
> I am not sure I understand you corretly, but there is a   standard
> formula for kurtosis, which is the 'exact' one I gave.  The   'approx'
> is a lot easier to use in practise and can go back much   further
> (longer period).  You can not implement the 'exact' in that   way in MS
> as far as I know, after about 20 periods you get some funny   error,
> so that is why there is an approx in the first place.
> 
> I   *think* dusant's formula measures the same thing, ie has the
> same idea in   it, but it is not the official kurtosis formula, if
> there is such as an   official formula.
> 
> Regards
> MG Ferreira
> TsaTsa EOD Programmer and   trading model builder
> http://tsatsaeod.ferra4models.com
> http://www.ferra4models.com
> 
> 
> ---   In equismetastock@xxxxxxxxxxxxxxx, "cardiop" <cardiop@xxxx>  
wrote:
> > 
> > MG Ferreira and Dusant...
> > 
> >   Friends...
> > You presented two formulate for Kurtosis... formulate   Them
> > different and take the different graphs. What she would   be
> > formulates it exata?
> > Thank you
> > Skill
> Yahoo! India Matrimony: Find your life partneronline.





------------------------ Yahoo! Groups Sponsor --------------------~--> 
What would our lives be like without music, dance, and theater?
Donate or volunteer in the arts today at Network for Good!
http://us.click.yahoo.com/Tcy2bD/SOnJAA/cosFAA/BefplB/TM
--------------------------------------------------------------------~-> 

 
Yahoo! Groups Links

<*> To visit your group on the web, go to:
    http://groups.yahoo.com/group/equismetastock/

<*> To unsubscribe from this group, send an email to:
    equismetastock-unsubscribe@xxxxxxxxxxxxxxx

<*> Your use of Yahoo! Groups is subject to:
    http://docs.yahoo.com/info/terms/