Yesterday, I have run my first simulations. I was dreaming of wild returns. I have been in for quite a disappointment. My super algorithm has at best a flat performance after accounting for the trading fee. The good news is that it at least, it doesn't lose money.
I have spent the vast majority of the night attempting to tweak and correct the flaws.
It generates a profit for about 50% of the times. and it loses the other half. And when it wins, it reaps just 1% profit. So this is basically peanuts for a $200 position....
It seems to be good in oscillating market but totally fails to recognize strong trending markets and take advantage of it. There is hope since I know what the problem is. I also know that my simulation runs with 1 second interval data. A live system could exit faster and better from positions due to real-time updates.
I'll have to return to the drawing table and refine my simplistic design.
Tonight is the night that I will be starting BASE.
I have spent the vast majority of the night attempting to tweak and correct the flaws.
It generates a profit for about 50% of the times. and it loses the other half. And when it wins, it reaps just 1% profit. So this is basically peanuts for a $200 position....
It seems to be good in oscillating market but totally fails to recognize strong trending markets and take advantage of it. There is hope since I know what the problem is. I also know that my simulation runs with 1 second interval data. A live system could exit faster and better from positions due to real-time updates.
I'll have to return to the drawing table and refine my simplistic design.
Tonight is the night that I will be starting BASE.
