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)
Yes I do, what would you like to know?
-I have been a member for approx 6 months. There is no trial period available, so deciding to fork out your hard earned cash is tough. I hardly ever buy anything trading related. Too many scams around.
-Is it legitimate? Yes absolutely.
-Does it work? I can't tell whether it will work for you as it depends what you feed it and how you use the software. It is a tool, nothing more. I will say that for me it has been a good investment.
I feed it tradestation data and for that it functions well.
Dave (the owner) is a good guy who will answer any questions you might have. He is also constantly working on improvements and listening to the feedback from the users. Many of the users are full time algo traders (although everyone keeps a low profile).
If you haven't read through the site and also the youtube videos- I suggest you do.
What I like most about Build Alpha is the tools are shamelessly trying to disprove any wishful thinking/ curve fitting.
Well I am an algo trader convert. I have learned some Easy Language but am still a beginner.
The ability for the software to build EL code is intriguing.
I'm struggling with the concept of using a Genetic Algorithm to assist in building prototype trading strategies. I feel like its pretty hard core data mining and very prone to curve fitting. I recognize that all strategies must be diligently tested to see if they are overfitted, but the GA concept seems so prone to the number one issue that plagues algo traders (overfitting). On the other hand it also seems to be a much more efficient way to test mulitple combinations of entries and exits. Maybe I am wrong and some more experienced guys can shed some light on this for me.
It seems the software has got some good testing built into it (walkforward, montecarlo, etc) but most good systems like Tradestation have the basics also. So it seems like your basically paying for a data mining software...
Like you I don't like buying any trading stuff because I don't trust anyone.
I'm no easylanguage pro but I don't believe that's a requirement anyway. I certainly wouldnt look at any software as a crutch for programming though, I use the software as a starting point and add to it as I see fit. There will always be some hard work required imho.
I have created a number of strategies by hand throughout the years on everything from metatrader 4 through to ninja and sierracharts. As I didn't have much success it seemed pointless to continue doing something which wasn't working.
So my thoughts were similar to yours about data mining and using GA. I guess you'll have to decide for yourself but what I have found is its all about your method.
Either process, whether its through a GA or a hand coded strategy can be curve fitted if you do it wrong. Falling in love with a strategy and tweaking it constantly and refining the equity curve... its a slippery slope
If you haven't already read it I highly recommend Kevin Davies book. It's very practical and discusses many of these issues with real life examples. After reading it decide what fits for you. Certainly don't rush into any big purchases.
I do not think Build Alpha is really datamining were you have the computor to find a pattern of OHLC that can be exploited. It has been fed 1000 ndīs entries and setups that is fixed and then are tested in diffrent combinations on the data you provide until it finds some combination of those entries and exits that would have worked in the past .
Actually i am for the concept and could see doing Gentetic evo and create a pool of millions of fixed entries and exits that were posetive on a a range of instruments and then use a GPU find the best combinations for the one at hand.
I purchased BuildAlpha as I was intrigued by the no coding required part since I'm not a programmer. While the software does a good job building and testing strategies from popular indicators very quickly. You really have to know python to get the customization that can bring it to the next level yourself. But the creator Dave is very helpful in every aspect.
All in all due to the price tag associated I would recommend it as an entry way. I was able to extract some good ideas off it and made a few profitable strategies. The other software I was interested in was the PAlab but that cost like 6x BA, maybe in the future I will try that one.
