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 really want to push amibroker to the limit and see what its capable of doing.
Has anyone tried anything crazy like every ticker over 100 yrs of data with some strategy it be interesting to see how quick and powerful you really can push it
Have people linked it to Matlab ?
Can you help answer these questions from other members on NexusFi?
People have used it with Python to create neural network.
I believe someone used MatLab as well.
About the suggestion of pushing AB
If you code right, using arrays, it will be as fast as possible.
But one speed limitation is the database size.
You can have the size you want but that will take unnecessary time to process.
There is no point of using a lot of years in a database.
As the markets change over time, the strategies should change as well.
In fact, that is the idea of WF (walk forward optimization).
A lot of users believe that store tick by tick on the database is useless.
On a high volume market like ES, it is just unnecessary data to be processed.
A 1-second database would be much smaller with the same accuracy.
What AB can do is much better than processing a century of data (Was there any market data back to 1916?).
It can receive data really fast, process it really fast as well as our code, and, take the decisions necessary.
To do that, it process only the data that matters.
If I am using a 50-period EMA, only the last 50 bars will be computed.
Or, AB can scan all the stock market (in a flash) to see which symbols are with a buy or sell signal, according to the setups programmed.
AB can do that as often as your data arrives (1 time a second, 30 times a minute...).
It is up to your internet connection and Data provider.
Those are the things that really matter to make money on the market.
And that is what AB do.
Is there a portfolio feature so you can a watch list weighted by volume and then which ever futures eg the ES is most liquid it can allocate the daily amount for trading capital?
You can balance or rebalance your portfolio by all futures in the Watch list, or only by the futures that had buy or sell signals, or any rule you would like (balance it only for the 8 most liquid future cause those are the ones I will trade).
If you google "amibroker portfolio rebalancing" it will come up with methods of doing that.
Some use AB CBT (custom backtest feature) to do it on the fly, day by day, week by week, you choose. I consider this a advanced feature.
Others use an exploration (analysis / explore) to generate the volume on a time range, balance the portfolio and backtest or enter the positions with this pre-balanced portfolio. I like this last better and this would be a basic to medium feature.
I still did not get what need do you have but he is very good.
For example, he can code your AFL in C++ so Amibroker would run it hundreds of time faster (if its complex).
Instead of taking 0.35ms, it would tack 0.035ms or so.
But, once again, you should get Amibroker, see how it works, see what would you like to do with it and how and, only, them try to make it.
