Though I'm not an algo trader, I've programed many simulation systems against random generated data and large sets of past real data (no difference in price movement) and the simulations tell it all, and as you said: nothing programmed works all the time, or even against different vehicles. In my opinion: it becomes far easier to just learn how to trade than to get hit with a series of algo losses, figure out what changed, test the changes, and if you have money left after the losses, go back in to run again. I'm not sure if AI will be of much help against real time decision making against random order flow type price action prevalent in the FX market in particular.
![ILLSLICK - KILLSHOT REMIX [FULL VERSION]](http://i.ytimg.com/vi/RIK8RrqIAzE/mqdefault.jpg)
![ละลาย (LALALYE) - DAOU PITTAYA ต้าห์อู๋ พิทยา [OFFICIAL MV]](http://i.ytimg.com/vi/wo6r9TgausI/mqdefault.jpg)
Though I'm not an algo trader, I've programed many simulation systems against random generated data and large sets of past real data (no difference in price movement) and the simulations tell it all, and as you said: nothing programmed works all the time, or even against different vehicles. In my opinion: it becomes far easier to just learn how to trade than to get hit with a series of algo losses, figure out what changed, test the changes, and if you have money left after the losses, go back in to run again. I'm not sure if AI will be of much help against real time decision making against random order flow type price action prevalent in the FX market in particular.
My style of trading. Tons of research and statisics. The problem though is the huge work load that can burn you out.
@Etienne - how many markets do your statergies need to work on, before you consider them "good"?
How often are you re-optimizing your automated systems?
Thanks E good stuff
Glad you enjoyed it!