I need some help interpreting the results of a study which looked at transaction records of US T-Bond Futures. The study can be found here - http://citeseer.nj.nec.com/cache/pa...SzpaperszSzKDD98.pdf/weigend98discovering.pdf Is anyone willing to take a look at this study and then field a few questions? This could be done here on the forum or privately. Thanks, nik
hi nik, i just looked onto it, has some cool points in it, however, the technique requires information regarding accounts / real behaviour of traders, which i (we ?) do not have, so i am not further interested in it.
Maybe that's why I can't figure out which criteria are used to derive the initial 7 clusters. I was sure it was because I don't have the math chops to figure it out The authors are interested in discovering "What trading styles and hidden regularities can statistical methods uncover in bond futures" and "What are the profit and risk characteristics for these styles?" These are the aspects of the study in which I am most interested.
i am sure these are for educational purposes... also, not to forget major points from this report are from 1992, markets changed considerably since then (outcry --> electronic move), i think it is save for us to forget this and continue own research ...