I have provided a list below that attempts to rank different ways to automate trading according to their degree of sophistication. That is, the list is meant to be ordered with the least capable systems at the bottom and the most capable systems at the top. I appreciate that the ordering of the systems on this list may be somewhat open to dispute. It is my intention to open this matter up to discussion, so that we all may learn. Please feel free to post any comments or disagreements. Completely Custom Application Custom Front End to a Broker's API TradeStation linked to another broker (for better commissions, different instruments, etc.) TradeStation WealthLab, TradeBold, TradeBullet, AmiBroker, NinjaTrader, etc.
Norm, how have you been getting on with your automated system development? Which route did you finally go. regards, Runningbear
Fully automated, great. Gain speed and abstain from unhelpful emotions. I think that in any automated system, the Master (you) must have an absolute authority on entry/exit decision at any time if wish to intervene. Imagine if you have a robot that can decide everything at any moment on its own and didn't listen to you. What kind of disaster could happen?
Tradestation, Tradebolt, and Wealthlab are all fine products. Tradestation / Tradebolt combo are great for e-minis and ecbot. WealthLab is great for writing programs but harder to automate. I don't have experience with any of the other products on your list so I will refrain from commenting.
This statement is quite strange unless it applies to a pre-version 8 release of Tradestation. Why would anyone use Tradebolt/Tradestation combo unless there was a substantial penalty from simply placing the orders thru Tradestation ?
Runningbear, I have a strategy that I am satisfied with. However, I am still trying to determine how to proceed with the implementation. The strategy is alive and well as written in EasyLanguare and running on TradeStation. I want to execute trades with IB because of their lower commissions and because they will let me actively trade in an IRA. I also want to develop a more autonomous program that I can provide to others. I have linked TradeStation to IB, so that I can run my strategy in TradeStation and execute with IB. However, I view this as an interim solution. My present strategy is very latency tolerant. Improvements to this strategy and other strategies that I want to develop are likely to be less latency tolerant. So, I want to take the next step toward a more autonomous system. I don't know yet what this will be. It may be a C++ program that interfaces to IB's API (and possible to other brokers). But, I am also exploring the use of FIX/CTCI and maybe even exchange membership. I am open to suggestions and would truly appreciate any comments. One of the purposes of my original post in this thread was to promote discussion of other possibilities. I am relatively new to automated trading and would like a better understanding of the available options. Norm
I have gone both the multi provider and tradestation only routes. I can say from first hand experience that the TS route made more sense. Current releases are very fast and stable. I autotrade without any issues for months at a time. Fewer moving parts make a big difference when things DO go wrong. Having to track the status of fills and cancellations alone during data surges and interruptions is a big deal. Ts does a good job so far. In a multi app solution it can be a big problem. unless you can program your own App, the cost of getting the job done to a satisfactory level would exceed any savings you could hope to achieve using IB vs TS. Also, the commission difference is really inconsequential unless you are trading large volume. In which case, upgrade to the F1 pricing plan (500 per month) and you'll get 1.66 a side for CME traded futures. You will be ahead after 225 RT or so. Obviously the lack IRA and lack of foreign markets are additional considerations unrelated to the apps themselves. One more consideration: Opportunity cost. While you are putting together your perfect platform, others may well be discovering the inefficiency that comprises your system's edge. Take advantage of something that works while it is still working. Best of luck to you.
Norm, if you are not a professional programmer (5 years experience)... You will never get beyond TradeStation... Any more likely than you will perform brain surgery... Or argue a case before the Supreme Court in the next 6 months. I've been working on a ** comprehensive ** Custom Front End... Using the IB API on an XP/VB/Excel platform for 6 months at 15-20 hours/week... And it's going well and will be done in another 3-6 months... But even with a Computer Science degree and 20 years experience... This is easily the most challenging project of my life. And the finished product will be perhaps 20% of my total proprietary trading systems. ** Robust ** real-time systems are a bitch to design, code, and debug. I do 150 trades/day with IB in a highly specialized part of the market... And as much as half the orders can be fully automated. So I'm really just "optimizing" my existing trading. Only down the line will I introduce new algo strategies... That may or may not work out. My ultimate goal 9-12 months out is to build a virtually stand-alone Bot trading a 200K account... That can be a "building block" for a Bot Army. Norm, I've been watching you spin your wheels here for weeks. I feel your frustration... But only a professional can build what you need... And no talented software engineer is gonna work cheap. rm+
The realtime version of AmiBroker (< $300) includes an Interactive Brokers realtime plugin for trading. There are a number of people that use it and I am now paper trading using a fully automtated a trading application(stocks and options) that trades based on programmed signals in AmiBroker to IB. For data you can use IQFEED, Esignal or even IB data feed itself. AmiBroker is very fast and has a bunch of stuff pre-built so I'm not sure of the advantage of using C++, Java or something else unless you have very unique needs.