Hi Matka,
It was recently brought to my attention that Van Tharp presented the results of some random entry experiments in his book "Trade Your Way to Financial Freedom". That led me to an interesting discussion on that topic, which branched out to discuss software random number generators in more detail. The conclusion seemed to be that the Mersenne Twister algorithm is a good choice for trading experiments.
That said, for my random entry experiments described in this thread I'm not too concerned with the theoretical quality of those entries. I'm more concerned with demonstrating what it's possible to achieve using apparently very "dumb" entries, as well as determining some sort of baseline from which to measure the "edge" provided by supposedly more "intelligent" entries.
Ah, I see what you mean. The trouble with MT4 in this context is that the platform itself is NOT open source. Barring the odd patch, getting bugs fixed and improving functionality is not easy to do. MetaTrader may well be popular, but that doesn't mean it's the best tool for the job. Just in case it might be of interest I've started a forum for discussion about fully open source trading platforms, starting with TradeLink.At the moment I am using MT4 with patches, not open source but fairly tested plus lots of other advantages due to it's popularity.
I guess we'll all have to start trading currency futures instead!I agree with you that ultimate test for backtester is comparison with live result tests, however sometimes it is not that easy due to non centralized nature of FX market and differences in quotations
Cheers,
Jim



Reply With Quote
