Hi, I am working on a hosted automated trading platform. The idea is to allow visual programming (no language to learn / no coding), back testing and operating using our platform. I am interested in what kind of buy/sell conditions most systems would be based on. We could support hundreds of technical indicators and price/volume conditions but in order to simplify I would like to focus on those conditions that matter to most users. Also I am not sure whether to focus on intra day strategies or mid- / longterm trading systems. Any comments are appreciated. Best, Fabian
What kind of users would be interested in a hosted trading system while also wanting a visual programming environment?
Good point. Also, would your platform be local software (installed), or cloud-based? Furthermore, would the user's IP be protected from, you? If so, how? I've seen one visual flow-style designer type platform that run on local machines. But it is not a hosted solution.
I think platforms like Ninja Traders, Strategy Desk, Wealth Lab etc. are too complicated for the average investor. Therefore these platforms have only a small audience. But the average investor who has no time to follow the market could benefit a lot from a systematic way to invest. Building, testing and running an automated trading strategy in less than 5 minutes would be my goal to open up automated trading to a bigger audience. This can be only achieved without software download (people at work cannot install a software locally). And hosting this in the cloud makes it scalable. For example we can backtest 5 years of data in less than 2 seconds. In the free version, all systems would be 'open source'. Free for anybody to see and improve. There will be as well the possibility to run a system as a black box. IP would remain with the user, but technically we have access to all systems. I don't think this is a problem. When Salesforce.com launched, people were concerned as well regarding IP and data access. But if something is easy to use and 'just works' the concerns regarding confidentiality are often 2nd priority. Anybody interested to test drive our Alpha in January?
If you can't program or hire someone full time to do it for you, the trader will not make it as a systematic trader. PERIOD. Good luck with churning newbs into losers during changing markets.
Your target audience is dumb and lazy, got it. Make that profoundly mentally impaired or so poor they can't afford their own computers. You mean just like I can do with every stand-alone tools? But I'm sure the cloud will make your 2 seconds appear so much faster. Ahh, here it comes. You want other people to write strategies and upload them to your site so that you can pick through them to find the best and trade them yourself or resell them. You wanted requirements? How's this, a big PayPal link and a button saying "Make Money Now". Anything more complex and your target audience may become confused. Best of luck.
Hi Fabian. I "host" my trading system myself and it is a lot of work; so there is surely a need however if you ask me I think that we are a long way away in maturity for hosting of a trading system to be anything like hosting of a web application. As for examples of what people do I suggest you look at books suchs as: New Trading Systems and Methods - http://www.amazon.com/New-Trading-Systems-Methods-Wiley/dp/047126847X This will give you a good grasp of the "technical" strategies that people apply. Also consider to look into more academic finance suchs as modern portfolio theory and portfollio rebalancing. I have heard of a couple of platforms that claims to do visual programming - i.e. this one: http://landing.saxobank.com/tradecommander/ I haven't seen it used in practice though.
You know why Tradestation is so popular? It's not because they offer such brilliant software or amazingly powerful technical indicators. It's because they offer years of historic intraday data. With TS, you get a complete package. Within MINUTES you can backtest intraday strategies on 1 minute data. With other software (Ninja, Neoticker, Multicharts etc.) the setup (importing data, setting up datasources, making sure new data is coming in etc.) takes days if not months. If you're looking to compete, start with data. If you get this right, you'd be lightyears ahead of 90% of your competing products.