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)
Backtesting and Research: MultiCharts or AmiBroker
I have both of the books "Quantitative Trading Systems" and "Introduction to AmiBroker".
They go from the beginner to AB to quite more advanced concepts.
As I mentioned when I first joined this subject, I am a professional programmer (or was) and I like to put my efforts into making systems that work not learning how to get around issues/limitations of the programming language.
It will be very interesting to see what the code looks like and how easy it is to read, when ArshT comes back with the AB equivalent to the very simple EasyLanguage example I gave.
Also, he has not answered the question about Intrabar testing.
It's amazing how you can turn a profitable system on a 5min chart into an ordinary system when you use a 1min intrabar test and then into a crap system when you use true tick data.
Same system, still using a 5min bar chart but looking inside each bar.
Just in case your not sure on what people are referring to in the above, as an example, on a 5min bar, say the open was 1/3 of the way up on the bar and the close was 1/3 down from the high of the bar. What came first, the high of the bar or the low of the bar? Unless you look and test at the intrabar action you can't tell. Multicharts allows you to drill down inside the bar action and test this.
This may be why MC is some what slower than AB. If AB is only testing at 5min level and no intrabar test and MC has been set to test down to the tick level inside the 5min bar, of course it will take longer but it is exactly as per real life action.
If others codes are like this, I might be willing to give it a go...
I follow you on the intrabar testing alright. Googling on this topic for AB reveals some difficulties, or that it is a lot of work. That was in 2009 though. Another discussion in 2010 found intrabar questions unanswered. Would be neat to see what the situation is now...
That said, for my current purposes, this would not seem to pose a problem for me. Either way, I'm finding this very educational and helpful.
True but where is the StopLoss? Is it at recent Low's. No it is a hard fixed value from the entry in points or %. Not a TA defined support area.
How many have you bought? It's not possible to scale out of the trade. After reaching a certain profit or if the TA StopLoss was going to be to far away, go for a smaller position size. I can't easily see how you can do any of this in AB but MC is a piece of cake.
ArshT,
I politely ask, perhaps you can show me sample AFL code I need to put into a System that will allow me to detect when I'm in a trade.
I just updated my Amibroker to the latest version and still cannot find a reference.
Using Multicharts as …
Buy 3, Sell 1 at a Profit target of 5 points, Sell 2 at a Profit Target of 10 points.
StopLoss set for the Low of the bar prior to the Entry.
You can do Intrabar testing too. I.e. simply use timeframe functions and expandfirst. The LAST bar close includes the actual LAST price regardless of timeframe. It was even answered by T.J. So please stop the misinformation. And as you already mentioned to get 100% accuracy simply use tick data. You wouldn't need CBT to use those functions. I haven't use custom backtester much yet (just added some additional metrics and code for adding slippage etc) so I don't know how to do it there. I haven't explored the complete list of capabilities of AB yet. Custom backtester is an environment within AB where you can design your own backtester or just add additional stuff. Just show me a software of the group "usual suspects" where you can do that. It opens the door to 100% flexibility. Just look at the links with examples provided by Tomasz.
Sorry but wrong again. Both were tested with same conditions in one timeframe backed by a MC user. If time permits and I should feel like then I could do a test with both using intrabar testing.
@Tony, if you can work with Excel you can also work with AFL. IMO, it is very simple to use and powerful (and yes AB with AFL/..../... is faster. It's widely known for being one of the fastest if not the fastest one with high stability and very few bugs). But again you aren't even forced to use AFL. And I don't think there is any software without a learning curve
But I'm not here to advertise AB. I just read TonyB's question followed by the misinformations by two users and wanted to give a reply. The first time I heard of this board was sometime last year. So my interest hasn't been enthusiastic so far (one reason, not having much time to lurk around every board on this planet :-) ).
Tony and I would like to see the AB version of the small sample system with a TA stoploss and two stage profit exit.
Could you demonstrate this please.