I am currently writing about using Machine Learning algorithms to discover Optimal Trading Rules. This work will be devided into several parts, the first of which is about developing an appropriate training set to use for the Machine Learning algorithm. I am working around an idea based on the use of Oracles in forecasting. I will put a draft of this paper on my author page at the SSRN when it is complete.  

I'm finally preparing to fully install my systems onto cloud computing platforms. This includes deleting the static IP number I have for my server rack on the farm in NJ — which may cause a little disconnectivity over the next week, but should substantially reduce maintenance going forward.  

I put the most up-to-date version in Google docs. If you follow this link you should be able to get to it. I will update this from time-to-time.  

My ancient Dell servers are decaying... odd that, since they're made of rocks and metal. I've been too busy to attend to it, but got this across to a new server a few days ago. Seems like I should do a regression of the performance of MF Global onto the dynamic trading risk factor. That will need a little work as I've moved my databases onto MySQL and there's still some teething troubles to go through.

[11/08/2011] And then squirrels ate through my power cables and I had to have the cables to the house repaired by Jersey Central Power & Light — and that work killed the cable internet which had to be redone by Comcast…

 

ZeroHedge has posted my updated chart plus commentary on the time series of the cumulative kurtosis of US three month treasury bills. For anybody interested in this work, you can access the raw data directly from the F.R.B. of St.Louis's “Fred” database.

The chart was created using the RATS time-series analysis program, and the code use is that below:

calendar(d) 1954 1 4
allocate %today()
set grid = %if(t.eq.1970:02:01.or.t.eq.1978:01:31.or.t.eq.1978:03:08.or.t.eq.1979:08:06.or.t.eq.1987:08:11.or.t.eq.2006:01:31,1,0)
set proart = %if(t.lt.2007:10:01,1,0)
set perart = %if(t.ge.2007:10:01.and.t.le.2008:11:30,1.,0.)
set secart = %if(t.gt.2008:11:30,1,0)
open data "DTB3.prn"
data(format=prn,organization=columns,skiplines=11) / value
close data
set yield = value
difference yield / dyield
set stdev = %na
set kurtosis = %na
do i=10,%allocend()
statistics(moments,noprint) dyield 1 i
set stdev i i = sqrt(%variance)
set kurtosis i i = %kurtosis
end do
set tseries = %if(%valid(stdev),t,%na)
compute tmin=fix(%minvalue(tseries)),tmax=fix(%maxvalue(tseries))
set stdev = stdev*100
spgraph(hfields=1,vfields=1,header='Cumulative Kurtosis of 3 Month Treasury Bills',footer=%dateandtime(),subheader='Daily Change in Yield: '+%datelabel(tmin)+' -- '+%datelabel(tmax))
graph(grid=grid,klabel=||'Execess Kurtosis','Daily St.Dev/bp','Bill Yield/%'||,key=loright,shade=perart) 2
# kurtosis
# stdev
compute lvpos=35.
grtext(entry=1960:9:1,y=lvpos,valign=bottom,align=center) 'McChesney Martin'
grtext(entry=1974:1:31,y=lvpos,valign=bottom,align=center) 'Burns'
grtext(entry=1978:11:21,y=lvpos,valign=bottom,align=center) 'Miller'
grtext(entry=1983:8:8,y=lvpos,valign=bottom,align=center) 'Volcker'
grtext(entry=1996:11:5,y=lvpos,valign=bottom,align=center) 'Greenspan'
grtext(entry=fix((2006:06:01+%today())/2),y=lvpos,valign=bottom,align=center) 'Bernanke'
spgraph(done)

n.b. Some textual translations are needed for the downloaded data to be directly readable by RATS (change of NA and date format).

 

This blog was down for about a week because Hurricane Irene caused my server's power supplies to get fried. Meanwhile, I am continuing to work on programming execution algorithms for our new alpha strategy.