Welcome to NexusFi: the best trading community on the planet, with over 150,000 members Sign Up Now for Free
Genuine reviews from real traders, not fake reviews from stealth vendors
Quality education from leading professional traders
We are a friendly, helpful, and positive community
We do not tolerate rude behavior, trolling, or vendors advertising in posts
We are here to help, just let us know what you need
You'll need to register in order to view the content of the threads and start contributing to our community. It's free for basic access, or support us by becoming an Elite Member -- see if you qualify for a discount below.
-- Big Mike, Site Administrator
(If you already have an account, login at the top of the page)
I'm considering using MC to trade equities and ETFs. I have a few questions about it with regard to MC and perhaps others who have used MC in equities environment can chime in here.
First big question I have that I want to discuss in this first post: I want to run a backtest against all symbols in the stock and ETF universe for research purposes initially. The reason is to get data/statistics on the backtests to see how well it performs overall across different stocks. Then I could datamine these results to determine a good filtration system (i.e., minimum price and/or minimum volume; cap on daily ATR, etc.) to apply for a real world trading system.
I think I read that MC can only be run against 100 symbols max?! Is that true even for backtesting? Or is that for only realtime order execution? I want to make sure there is no artificial limitation like that (especially for backtesting where response time isn't an issue when just collecting data to study; e.g., AmiBroker can backtest against entire stock universes or one can specify watchlists/subsets of the universe to run a backtest against - its up to the end user to choose). Typical stock universes run into the thousands of symbols (e.g., Russel 2000 itself is ~ 2000 symbols; the Nasdaq exchange has over 2800 listed companies as of beginning of 2011). I really want to use MC to backtest against thousands of symbols. Can I do this? If not, why not (more a question for MultiCharts engineers)?
I have other questions too regarding the MC platform but the above is my first major concern that I'd like to tackle with regard to MC for equities trading.
Cheers,
JD
Can you help answer these questions from other members on NexusFi?
I'm currently using MultiCharts for essentially the same purpose. I haven't actually traded using it yet, but I'm using it to scan for securities and backtest various strategies. I'm using it to backtest up to 100 symbols at a time, like all of the symbols on the Nasdaq 100, to identify the best ones according to my criteria (e.g., the best trends).
Unfortunately, the 100-symbol limit appears to be accurate. That's not a limitation on trading, but on backtesting, for their regular version. (Trading appears to be limited by how many charts you can have open at any given moment, without it choking.) They also sell a "pro" version that allows "over 500" symbols, but that costs $10,000 (which is a bit cheesy in my book).
The good news is that according to a recent announcement in their forum, the next version of MultiCharts will allow more than 100 symbols. I don't know if that's referring to Version 7 (in beta) or Version 8 (not announced). Also, it's not clear if that means "no limitations on symbols," or just a higher cap.
Part of the problem is that the current 32-bit version is limited by the amount of accessible memory. The portfolio backtester sometimes chokes up on me when I'm doing large backtests, like 3000 daily bars with 100 symbols. However, the Version 7 beta seems more stable than Version 6 in this regard. Also, they've indicated that they're working on a 64-bit version, hopefully for later this year.
By the way, I'd appreciate learning more about the AmiBroker backtesting. The website for AmiBroker looks rather amateurish, which makes it hard for me to take the software seriously. (I know that I shouldn't judge a book by its cover...) However, if it's a viable alternative to MultiCharts, that would be good to know.
I've been using Amibroker for backtesting for about 6 months now and love it. Its simple and straightforward to use for this and the community is great. There hasn't been much in the way of new development that i've seen but for a simple and cheap way to backtest the universe of instruments, Amibroker is a good option.
Thanks, that's great to know! I'll give it a shot.
For the sake of asking, have you done any automated trading with Amibroker or another program? My goal is to launch an autotrading strategy soon using Multicharts with Interactive Brokers. However, I'll probably need to buy another Multicharts license to do so.
I'm paranoid about the issue of reliability -- I don't want to have to babysit the system constantly. (It will be running on a dedicated server at a datacenter.) Multicharts seems to have a good reputation for reliability, but I'm willing to consider alternatives like Amibroker.
I've not done any automated trading with any platforms. I do know of a few people who used amibroker for automated system trading but most people i know use Multicharts, Tradestation or esignal. I've never heard of any issues at all with any of these three.
I wouldn't personally trust AmiBroker for auto trading despite it having add-on to send orders to IB (its called IB-Controller or IBc). AmiBroker is really a great backtesting engine and is great for doing research type work (like said before - it can handle backtests against complete stock universes or dynamic filters applied to such universes; it takes into account how much equity you have left in your account from trade to trade; it takes into account liquidity/volume of a given stock and not take on more than 10% volume's bar for bar you trade on; etc.).
However, AmiBroker really doesn't have a strong footing with auto trading, IMO. I suggested to Tomasz (developer of AmiBroker) in Las Vegas at a AmiBroker user conference few years back that he should move more into this space. But he seems nervous about liability purposes for some reason so hasn't really put a lot of effort into making software that would be key for autotrading (e.g., position tracking, position syncing between your strategy and broker's position set, etc.).
Worden Stockfinder is the best client side stock scanner I have found. I use MC for macro scans but if I want to scan the entire US equity market on a near real time basis I go to Stockfinder for that.