Hi,
In many finance research application decile (or quintile or quartile)
control is a commonly used procedure.
A simple example would be, say you want to assess the abnormal stock price
performance of a sample of companies and say you think the size of the
company (measured by say total asset value) is a important determinant of
stock price performance.
With the decile control approach, You take all the firms in the market
(excluding your sample) and sort them on the basis of the value of assets
and split them into deciles.
Then you allocate each of your sample company to the appropriate size decile
and compare the performance of the company with the average performance of
the decile companies.
Sometimes when you have several variables on the basis of which to form
deciles or quintile etc. the procedure becomes problematic.
I had a look at the manuals and the array of methods available is a bit
intimidating.
Could I use kmeans clustering as an alternative approach to decile control
In the first instance I intend to use bookvalue to marketvalue ratio and
the market value of the company to form the clusters.
Regards
rajesh
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
