These charts look really cool and I think I see what's going on, but how would you use this for the scenarios you describe such as trading?
Thank you for your kind words.
At the moment we're just experimenting with BTC of course, but the goal of our models is to price the target instrument better than the market. If we can price the instrument better than the market, particularly during volatile periods, we may then enter and exit our trades with far greater certainty in outcome.
http://blogdotquantsigdotnet.files.wordpress.com/2013/04/picture-161.pngThe above BTCEUR (Intermarket) model demonstrates good qualities such as the actual instrument returning to output (09:00,12:00,14:00,18:00), a near-oscillation within diverg, with diverg leading forward momentum(08:00-10:00,12:00-14:40,15:00-18:00). In effect, the model identified exploitable overbought and oversold conditions multiple times during the observation period.
We have a guide available which lightly explains our platform, what the models are, their states, etc.
http://www.quantsig.net/guide.html (chrome/firefox/opera)