Does anyone have experience with Black Boxes?
- Thread starter jnhighland
- Start date
StrategyRunner
Strategy Runner, Ltd.
Many people use the term black box has if it was a mysterious or magical item that simply runs automatically and makes money (hopefully), the simple and less glamorous fact is that a black box is an algorithm
It takes that technical analysis that traders and combines them with rules.
if X happens then do Y etc.
Strategy Runner over the years as written thousands of Strategies (back boxesâ¦) using out API (C++ based)
Once we write the Systems we allow you to back test it on Tick data (up to 5 years back) and do your out of sample testing on our simulation server. Once you are happy with the results we deploy your systems on our servers (located in the FCM facility) were it runs automatically (you can monitor and override it at anytime).
Our solution is used by many CTA's , Funds and privet trader that understand the advantage of this solution.
Send us an email for more info
info@strategyrunner.com
It takes that technical analysis that traders and combines them with rules.
if X happens then do Y etc.
Strategy Runner over the years as written thousands of Strategies (back boxesâ¦) using out API (C++ based)
Once we write the Systems we allow you to back test it on Tick data (up to 5 years back) and do your out of sample testing on our simulation server. Once you are happy with the results we deploy your systems on our servers (located in the FCM facility) were it runs automatically (you can monitor and override it at anytime).
Our solution is used by many CTA's , Funds and privet trader that understand the advantage of this solution.
Send us an email for more info
info@strategyrunner.com
I will be interested in such system:
1-Money in my account
2-System developer can and should trade it
3-System to be adaptive to market, not fitted
4-fundamentals taken into account also.
5-Important :System developer should be trading same
strategy, and you should be able to see what size he is
trading with, that means if you lose money, he also loses
money in his personal account, So he can't put you in risky
trades for more aggressive return.
6-You can stop trading anytime, if money goes below Max DD.
7-You decide how many contracts to be traded
8-You pay % from you profits, and no lease or management fees
Now, if anyone can offer such thing, sure he can find
lot of traders to invest with him.
1-Money in my account
2-System developer can and should trade it
3-System to be adaptive to market, not fitted
4-fundamentals taken into account also.
5-Important :System developer should be trading same
strategy, and you should be able to see what size he is
trading with, that means if you lose money, he also loses
money in his personal account, So he can't put you in risky
trades for more aggressive return.
6-You can stop trading anytime, if money goes below Max DD.
7-You decide how many contracts to be traded
8-You pay % from you profits, and no lease or management fees
Now, if anyone can offer such thing, sure he can find
lot of traders to invest with him.
I have no experience with black boxes but if I were a programer looking to write a program to trade the market for me I would look at scalping between futures prices and futures spread prices. I have done this manually with a bit of success but it is very boring and time consuming.
However I would stay away from anything like moving average crossovers, RSI, fibs or anything based on T.A. If you do this is how it will likely work out; first backtesting with excellent results, then go live with a little bit of money and make some small profits, then you leverage up the system before a an earnings warning or oil price spike and you lose big.
However I would stay away from anything like moving average crossovers, RSI, fibs or anything based on T.A. If you do this is how it will likely work out; first backtesting with excellent results, then go live with a little bit of money and make some small profits, then you leverage up the system before a an earnings warning or oil price spike and you lose big.
I've run my strategies on both TradeStation and Schonfeld. TradeStation was great for backtesting and Schonfeld was best for speed and executions.
Ahh, yes indeed, an often asked question. The answer is clearly obvious, yet (and this is gonna piss a few ETrs off - but like I give a 4x) as per usual the vendors give the same lame-arsed excused, whilst we know the real answer.Quote from cvds16:
Why would anyone who can develop this need outside investors ?
I don't understand why so many traders refuse to believe that black box systems can be profitable, is it the fear of letting a program make decisions? I totally agree that the majority of profitable black box systems will never outperform the same percentage of profitable manual traders but it depends on the complexity of the enigne.
If you were to write a algorithm that purely trades off indicators I agree that in the long term you're more likely to lose money. If you were able to code in logic to take into account x number of other factors (I'm not talking about going into monte carlo simulations) then a profitable system is very likely. I've written such an application that trades FX by itself and whilst still in development and some of the logic is crude, its performance is consistantly profitable (nothing stellar). I've spent plenty of time coding in logic to decipher price action into something meaningful rather tha relying on indicators and prevent the app from trading around certain times (figures)
I'll explain it to you.
Some people indeed have their heads in the right place and realize that the term 'BLACK BOX' is perfectly meaningless. It's of the same kind like HOLY GRAIL, TREND, MONEY MANAGEMENT and so on.
Of course people not belonging to the above sparse class but dreaming of themselves as being (profitable) traders are very hot on this kind of pixie dust.
Hope this helps,
nononsense