Subscriber Question
Your system is doing amazing well. The system seems to capture most the trend at the right time, meanwhile other systems on C2 is having hard time to find trends. I wonder if you can offer smaller systems with same algorithm that can fit account size 50k and 100k. Although I do realize that smaller system may not offer the performance that the current system do since it will do a smaller set of commodities and therefore lack diversification. But most investors have 50k or 100k to trade. 200k is probably hard to come up with. Please let me know if this is feasible in the future. Maybe you can run a back test, say 50k system, from selected commodities for the bigger system and see how it compare to the big system. If the smaller system still have comparable result, then it would allow customers to build their capital and when they have enough capital, they can switch to the bigger system.
Answer
Thanks for the kind words. System performed quite well in real time simulation over the last six month, but nothing like now. These are special circumstances and will unlikely to repeat to the same magnitude. It will perform to a degree of 30-50% compared to current performance in normal volatility.
There are two characteristics which make it so expensive to trade. Relatively large potential losers, up to $2000-2500 each and large set of markets. With today's margins, when fully invested it requires around 130K just for margin deposit alone. Add to it 20% potential draw-down at 40K and you have 170K to trade minimum so you can sleep well at night. When I was designing the system, I had maximum performance in mind. Apparently, it comes with the cost.
If you download CSV file from C2 with all the trades and sort it by the markets, diversification issue becomes apparent. Some markets are still expenses while selected few drive systems positive performance. It takes some markets 6-12 month to become profitable. If I knew how to isolate those performing markets or performing periods, I would make it a part of selection algorithms. I do not. I use exact same logic on all the markets. I decided to publish Basket Futures I because I realized that I could not trade it myself. Hopefully there are well capitalized traders out there who can. I agree with your idea, even for my own use. However, at this moment I simply do not have time to go through the exercise.
So, you can attempt this for yourself. Get list of markets from CSV file, find your broker margin for each market x volume. Group markets by Group /metals, grains, energy, currency, softs, equity index, interest rates, meats/. Pick two markets in each group. This will give you subset of 16 markets and lower your margin requirements to 50-90K depending on selection. Turn auto-trading on for these markets. Test this in simulation for a month, see how it does.
When I have time to do the work, I will attempt to construct smaller portfolio. There are some technology issues I have to solve to do that. That's why I do not have a back test portfolio results for the BF I at the moment. It is a matter of putting in the hours. I will let subscribers know as soon as something new becomes available on this matter.
Marat
|
