Working automated system with backtesting

J

jimmy

Hi, I have just joined the forum and am looking for expert traders who may be interested to collaborate:

I have a very strong background in technology and have created an automated trading system. It has a sophisticated back testing component in addition to order routing, and I 've been using it successfully for about half a year. I am using my own ideas, and have been staying profitable. However, the efficiency is not as high as I would like it to be.

In case you are interested to share ideas and possibly collaborate, I can certainly provide more details. The only conditions that I would be looking for are as follows:
- you trade for a living and are able to make money on trading;
- you are looking to increase your gains by applying your methods to a much larger number of stocks, which is easy for a program but impossible to a human.

Please send me a private message if you answer positively to both questions and would be interested to continue this conversation.

Thanks,
jimmy
 
Where do you live, and are you affiliated with a trading firm?

Are you interested in longer-term systems or short term systems?

What kind of drawdown is exceptable in your trading?

I have a few systems that make money but they do have drawdowns of 35% - when the markets correct / crash etc.

If your interested pm me.

Eric
 
I live in NYC and trade a prop account.

My system trades about 70-100 listed stocks a day, executing 200-400 positions a day fully automatically. It's pure day trading, I don't hold onto positions. Technically I can run a few hundred stocks at a time, but I would like to raise my average p&l first.

So, if anyone is interested and can answer affirmatively to the 2 questions in my initial post, please PM me.

jimmy
 
I'd be interested in 'collaborating' but would like to know what each party would be expected to contribute? In other words, what would traders involved in this effort effectively gain? Each others' techniques/systems?
 
That's a good question, and I guess the mechanism would depend on a particular situation. My backtesting system takes in complete set of market data from NYSE, up to the millisecond; it allows me to validate different ideas on a large array of real data. My execution system performs hundreds of trades a day and can perform thousands without breaking a sweat. Over time, I have made the execution pretty intelligent, so I don't have to be glued to the screen myself.

Having said that, I personally believe that a good day trader will usually beat a program in a trade; but the program can beat him ultimately on volume, as it may follow hundreds of names. So a working method that consistently makes you money in manual trading can produce multifold profits if implemented in such a system.

jimmy
 
There is a huge difference between being profitable in backtesting and in real trading. It is possible to be making virtual megabucks in backtesting and never make a real dollar. To make a distinction, I put the word "Working" in the thread subject.
 
Jimmy,

A couple questions may help others see if they want to collaborate:

1. What is your execution system programmed in?

2. Does it go through a broker API, FIX, or custom I/F?

3. Who is your broker?

TIA

C
 
The system goes through a broker API. Your other questions would be more suited for a private communication, with the 2 conditions set above in the thread.
 
You must log in or register to reply here.
Back
Top