Automated Trading Software

As I've said to TS before: you need an outsource skilled team with some technology base and code libraries. I do have all of that: the platform framework, the team, since 2002 we've got lots of perks, 5000 Hz HFT sounds good? And it's offshore.
If it's a matter of budget or intellectual property issues - we'll handle properly, please PM
 
Not using any platform, had planned to make our own custom one, but it never got completed. Have a partial platform which we wish to complete.

It takes a lot of time and human efforts to make a trading platform.It's rather interesting to look at what you've already done.
 
It takes a lot of time and human efforts to make a trading platform.It's rather interesting to look at what you've already done.

Yes u are right. The platform idea works well with my level of knowledge. It will allow me to create my own functions etc, without the need for a programmer, just some basic understanding in a simple language.
 
Not using any platform, had planned to make our own custom one, but it never got completed. Have a partial platform which we wish to complete.

That's ok, we spent 5 years to make a good framework.

You are looking for a miracle - getting such a skilled partner with commitment... What about getting a team, where knowledge is shared, and the environment is stable? Make a prototype and get some ceed cash. We are currently into 4 or 5 such startups. Check etna robot framework.
Think wider and don't waste time seeking for unreliable solutions.
Good luck anyway!
 
Last edited:
Ok, that now puts me safely above 15, just sent you a PM.

How is it going with Zorro Trader? I just went to to their site to see what that's about. Seems to me it's nothing more than TS / MT4 / Ninja Trader.
I'm skeptical about the money making aspect of this software. See their FAQ, it has many contradictory statements.

My background:
I've been a programmer for DECADES now and turned into a real trader in the last couple of years (previous 8 years of learning, i.e. buying expensive "systems", constructing my own systems, trials and errors, and losing $$$$$$).
I write my own software using VB.net (easiest) to gather and store data from IB (or eSignal previously), do needed calculations and place orders. I'm a currency day-trader, so my software GUI is geared towards fast executions with IB.

My assessment of your project:
Need to chart (very complex and time consuming) doesn't have to be custom-made. Their already exists good charting software.
Why do you need porfolio manager? Do you mean you want your software to track the order executions and store them in the database?
What do you mean by order manager?
Your true need, seems to me, is to be able to place orders via FIX or broker API after doing whatever necessary calculations. For most people, scripts integrated with, say Ninja Trader, should accomplish that. Why do you need your own custom program? I'm not saying you don't need it, because I certainly don't use EA scripts to execute my orders.
 
People developing serious automated trading systems probably won't see "MT4" as similar to TS and Ninja Trader. Zorro plays again in a different league. Admittedly, they all are software and have to do with trading.

Anyway, as I wrote the FAQ that you mentioned, I'm always interested to learn what could be improved - which statements do you think are contradictory?

"Employed professional traders can achieve an average annual profit of about 3% above market* - not with 'trading skills', but by following strict rules and discipline. Because trading is not based on real knowledge, it can not be taught or learned beyond elementary rules. It's unlikely for someone to become a 'better trader' over time, no matter how much money he invests in trading books and seminars."

"Still, trading machines are no better than the strategies they're programmed to follow. So far only large institutions, such as hedge funds or investment banks, have been highly successful with machine trading. Their high-speed systems exploit shortlived inefficiencies and minimal price differences between markets."

The 2 above statements are consistent - altho I disagree.
However, the above statements are contradicted, IMO, with claims such as
"Some of Zorro's included strategies can double your capital in about 4 months, and have confirmed that performance in one year real trading so far.".
BTW, I think we would all like to see that confirmation. Even then, many system can double your capital in a short time, with luck and excessive risk taking, just like gambling.

There are othere statements that are made that make me more skeptical:
"Trading requires capital. You'll need at least $200, better $2000 to start."
It would be crazy to start trading with $200. Even $2000, altho it may be possible with lots of luck.

"Trading strategies tend to expire due to permanent market changes."
Only if the strategy is not robust. The market is either in a trend or in consolidation. 2 states you will be able to count on permanently.

"Zorro's artificial intelligence is based on a simple neural network - a perceptron - and a decision tree learning algorithm. Additionally, Zorro offers several advanced analyzing tools for deciding when to buy and when to sell: Fréchet pattern detection, price cycle detection, frequency filters, correlation, polyonomial regression, and statistical arbitrage. Zorro learns while trading: it can regularly improve its strategies by re-optimizing parameters and capital assignment due to the current market situation."
I know a brilliant "rocket science, top of the profession" engineer would-be trader who has gone bust applying the above-like systems. To me, it's just buzz words. Also, isn't implying that these tools work, contradictory to your first 2 stmts above? And if they don't work, how is your software any better than Ninja scripts or TS?
I do believe that you can create a formula/system that can make money, just not with the above tools you mentioned - except statistical arbitrage - and then my skepticism level goes up again - how in the world would a normal private retailer do statistical arbitrage with your software - highly doubtful.

It appears that FXCM is currently the only "real" means to get historical data. That limits this platform severely.

Anyways, at least it appears that you are not charging for your software, but neither is Ninja. I want to be clear - I'm not putting down your software. You are promoting your free platform so you can sell custom programming. That's a reasonable service for traders who can't or won't for themselves (not withstanding whether your rate is reasonable or not). Altho that leads to my belief that if one is to dedicate oneself to trading, one better learn to do some basic scripts - only if to gain real insight to actual indicators/formulas/systems - and also to save valuable trading capital.

Have a great day.
 
Last edited:
This system is in life trading since September 2011. All parameters are open and the users can test the systems under all external conditions. However, I agree to you that life trading is not a method to properly test a strategy, and certainly no method to predict future performance.

What is "life" trading? Live trading? If this is real live trading results, I would look closer into your system.

This statement makes no sense to me. There is no such thing as a needed capital "to start trading". The capital you need depends on the drawdown and required margin of the used system, and both can be determined with WFA or Montecarlo simulation. For Zorro's two included systems it's $200 at minimum settings, for your own trading system you might need $5000 or even more.

True. I am thinking in terms of trading for a living or even making decent $ to compensate for the efforts put into trading.

fixed market states exist only in the imagination of traders
This really depends on the time-frame. For example, consolidation in 5-min tf could also be a trend in 60-min tf.

Perceptrons and regression trees are not buzz words, but existing algorithms
You misunderstand. I just meant say that they are big sounding words which, to me, are not relevant in trading. I like the KISS principle, especially in trading.
 
Top