5 Year Backtest on my Algo, page-8

Hi Tim and thanks to the others who commented or gave the post a thumbs up. I appreciate your interest in the model. Having no experience in trading models I don't know if it unusually good or nothing special.
It's probably a bit rude of me to post interesting model results and then not share the joy but I'm keeping the model close to my chest at least for the time being. It is about two months old right now from when I noticed the quirk in the maths that gives the results and I have just started working with live numbers (on a play account with a CFD mob so they can do the P/L maths for me). It sends me emails when I should change positions.

I suppose my motivation in posting it is to gauge its performance relative to what others have achieved and assuming it is encouraging, to let others know it's not impossible to create profitable algorithms. It doesn't use any high speed trading, it may hold a position for as little as a day or as long as weeks. I actually don't look at charts at all (no going cross-eyed or pissing off the boss). It's evaluating my positions once every 120 seconds time while doing my day job as an Engineer. I use Webiress feeding the XL sheet running a VBA script. It's a bit clunky but in this phase I am happy use it to see if the model merits further development. A quick look from 16 Oct to 30 Jan gives an average of 16% across the six stocks I mentioned in the first post and the rise is reasonably linear not some crazy saw-tooth which bankrupts you three times and suddenly gets lucky in the last week. I should add that 16% is trading long and short positions so only going long on stocks would probably be about half that.

I'll be back in the coming weeks/months with more information and real-world testing results. I am not against the idea of sending out signals to others in the future on a mailing list but I have a bit of work to do first.

Regards

Monte