Looks like I finally made it!

 

Well... after about 9 months of intensive programming it finally looks like I've succeeded with a signalling system that really works.

This "unreal" graph is a backtest result. But after two weeks of forward testing on demo the preliminary results are fully inline with the

back tests... getting about 10-15% pr day with very minimal risk.

But getting results like this means I'll have to throw like 20-50 Lot50 orders at the broker within the space of one or two minutes!

(20-50 Open orders, and later the same number of Close Orders).

Anyone know if a broker, like Alpari, will accept such order bursts inside one or two minutes ?


Link to Tester Graph:

http://home.online.no/~eikarlse/temp/TesterGraph.jpg

 

why not just ask them?

 

hi - it looks very unrealistic for real execution. with this volume you would need to use an ECN broker and not one with a trading desk.

 
May be it is a stupid question but are you sure YOU HAVE TO SEND 20 TO 50 ORDERS WITHIN A MINUTE OR 2? Can you share a little bit more of what you do? It also will be interesting to know what kind of strategy or idea or system you started with to get such nice results.
 

> But after two weeks of forward testing on demo the preliminary results are fully inline with the back tests.

As they should be

The challenge is forward running on live - hope it goes well

-BB-

 
Jacques366:

Yes, this is kind of a short term strategy and timing is critical... can't wait for 10 minutes to open or close... train may be gone then.

The reason for the large number of orders is simple: after just 2-3 weeks of trading the order size is up at

50 lots, which is the typical limit of most brokers (at least for Alpari). From then on, in order to keep the risk up,

multiple 50 lot orders will have to be 'fired at once'. At the most extreme it would simultaneously fire off

up to 50 50lot orders... which would emulate one 2500lot order.

I have since revised my strategy a bit by limiting the max number of orders to about 8 (equal to one 400lot order),

and obviously this will result in steadily decreasing risk as the account grows.

I figure a 99.8% hit rate with 400lots (typical daily profit around $50K) is going to be more than profitable enough for my needs...

 
DayTrader wrote >>
Oh ok, it's clearer now.
Like fbj said, I think that is something you have to discuss with your broker. They may have other types of account to allow such an activity like corporate ones (and it is not a big deal to register a company).
Another solution is reducing your expectation as you have done and run your program on a few accounts to multiply those 'little' profits.
Bravo again for your impressive results and Good luck for the next steps.
Reason: