Skip to content
Prev 1274 / 15274 Next

coskew/cokurtosis test

Brian,

I probably should clarify my view of co-skew and co-kurt.  I view these 
in multivariate distribution setting.  I googled "multivariate skewness" 
and get 11,000 hits with the first 20-30 providing good references.  
Several reference a paper from 1970 that defines Mult-skew as

or

I am still reviewing these.

The references for co-skew and co-kurt that I can find and I think you 
are trying to implement view these as the skew beta and kurt beta in the 
higher moment CAPM, which would be the multi-skew and multi-kurt 
measures scaled by the market variance.

In my example, I use the covariance of all assets to scale your co-skew 
function results to get the multi-variate skewness measures.  I am going 
to test my quick and dirty calculation against the above formulas to 
validity.

Hope this helps
Joe
brian at braverock.com wrote:
r
-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://stat.ethz.ch/pipermail/r-sig-finance/attachments/20070219/2779fe9c/attachment.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: moz-screenshot.jpg
Type: image/jpeg
Size: 2987 bytes
Desc: not available
Url : https://stat.ethz.ch/pipermail/r-sig-finance/attachments/20070219/2779fe9c/attachment.jpg 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: moz-screenshot-1.jpg
Type: image/jpeg
Size: 6051 bytes
Desc: not available
Url : https://stat.ethz.ch/pipermail/r-sig-finance/attachments/20070219/2779fe9c/attachment-0001.jpg