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)
Bob, you said you haven't done any development of automated strategies. What about you, Karen? If so, how did it go for you and how was the learning curve?
I don't do automated trading, but I write my own indicators, and my job for a bunch of years was as a programmer. My experience with the coding I have done on SC is that you would be well-advised to know one of the C languages (their stuff is all C++), then to dig into the documentation, which is not that simple, and to have some patience. If you've coded for a living you've done this a hundred times as you had to learn something new, and this is the same. It did take some time, and the learning curve was not exactly gentle.
Now that I've got the hang of it, it's quick and easy.
Again, I never wrote anything that did the executions for me, but this was my experience, purely in terms of the coding.
Bob.
When one door closes, another opens.
-- Cervantes, Don Quixote
Thanks again, Bob. That was very helpful. I haven't written anything in C (I don't think C# counts) in a long time, but I am capable of re-learning it. Or, I can ask you for help
I will download it this weekend (there's a free trial) and see how it works.
Day Trade; What a boring day! Trade worked. Not sure I was ever under water, but it just putzed along most of the day. Did the now-standard 3 units, and the most I was ever up was 1800, but there was no excitement. No big runs where I thought it might be a great day (visions of making up for yesterday in one day are generally in my mind when I have a losing day), but nothing strenuously bad either. I actually walked away from my computer for over an hour on two separate occasions just to combat the boredom. Thought about my trade quite a bit last night instead of sleeping, hence I was too tired to really do anything constructive with the time I had available. So that's disappointing because it can be hard to find time when you want it - today I had it and couldn't really make use of it. Up about 1495
Non Day Trade: Nothing done. Probably should have looked at option premiums for covered Calls, but the last few times it hasn't been worth it, and today I just flat out forgot.
Day Trade: A very nice day. Trade did almost nothing for the longest time. Several hours in, even after the 3rd leg was installed, I was only up 200 bucks. Then a very nice spurt took it to +1800 in just a few minutes, followed by a couple more hours of nothing. And then BAM! another big spurt. I was up 2800 and decided to call it a good day and a good week. Got very lucky on the exit as I closed out a couple longs and then the market took a big dump. It could have happened the other way, but on a trendy day I exit the side that has been losing first. Sometimes it reverses just after I do so, and that's irritating. Today it didn't. I exited the Longs, rode the short wave a little (but nowhere near as much as I could have if you had told me just where the low would be) and exited +3323 for the day. Good day and good week. And unless I really really screw up Monday, it will be another good month.
I did once in an older platform called Neoticker. Hired a good coder for the base of it as I can't do anything like C# etc. I think the toughest thing is just adapting money management to market conditions. Market tempo can vary a lot between 3 months, 6 months, multi-years, decades, etc. Every automation strategy i've personally seen needs lots of hand-holding. I'm not doing automated now, but if I did again in the future, which is a goal of mine, I would do grey-box style. I think automating just parts of your technique is the way to go, to help reduce stress and subjective decision making, if there is any. But this is all personal opinion of course.
Although I'll admit I have no experience with automation and therefore don't know what I'm talking about (not the first or the only time ), this makes a lot of sense to me, too.
If a trader were trading on a completely "discretionary" basis, but had a set of rules that they did not ever violate, this would be similar to this "grey-box" idea. Run with the rules, but have a discretionary override. Personally, if I did try automation, I would be much more comfortable with something like this, at least at first.
A good part of the appeal of automation is to take the trader's individual reactions out of the equation, but my concern would be that you then are assuming that the market will always, or near-enough to always, conform to the analysis and testing that went into your automated system design, which I think is kind of questionable.
Of course, another part of the appeal of total automation is that you would (hypothetically) just let the thing run and it would make you rich.
This does appeal to me greatly.
Sorry, @jba1962, I know that this kind of semi-automation is not what you are interested in. Just some thoughts I had, sparked by the discussion.
Bob.
When one door closes, another opens.
-- Cervantes, Don Quixote
I haven't automated yet, but if you want to fully go automation way, you will be in for a ride. I've a friend here in my country who had gone trough that process (little out of touch) but he had to spend almost a year to get everything coded and working. Note that he didn't just code his strategy, but also API to exchange/broker platform.
It was LOT of work, and he needed to hire team of coders for API development, and it was especially difficult to have working communication between trader/mind behind strategy and just coder for hire.
He utilized VB coding for API, strategy was coded in AFL and platform used was Ami, I'm not sure but he also had to take some service from exchange directly. This was in addition to have his servers up and running in almost 24*7 condition. That's all he was willing to tell me at that moment. I hope it helped you in some way
If you want partial automation though, I think its not that difficult to achieve if you can utilize broker supported API and use your coded strategy to trade with. As long broker API supports your code platform, it should run without problem. I assume its more smooth process at your place. If API supports, one can even utilize excel to automate orders, but that does scream little shaky ground.
Honestly, automation is bit out of my ball park atm, I only keep gathering info but Good luck to you brother!
FYI, this would be a very elaborate method of automating trading. It might be necessary for some trading strategies, especially that involved high frequency trading, but if you just need to automate a strategy that could be executed by you on your PC manually, nothing like this would be necessary.
Most platforms have a programming language that will simply execute trades from your PC, in the platform itself, as if you were doing them manually, using some criteria that you have established in the code. Trades go to the broker the same way any trade would. Nothing else is required.
If you really wanted to, you could put the platform on a VPS (virtual private server) that can be rented from many providers. Then you would access the platform remotely from your PC, but it would run on the server and not use your internet connection, power line, etc. Most traders who automate don't seem to bother.
The point is that you don't have to build all that infrastructure if you don't need something extraordinary. Now, the hard part is always going to be to have a strategy that will actually work . Then you need to code it, test it, curse when the code doesn't work right, fix it, repeat many times, then when the code works, test it over and over again until the strategy works, etc. Then go live.
Hopefully, then start cashing checks. I'm not saying it's easy, and I haven't done it (nor do I want to), but it's not necessarily such a huge undertaking, unless, again, you are doing something with extraordinary requirements.
Apologies once again to @jba1962 for filling up his thread.
Bob.
When one door closes, another opens.
-- Cervantes, Don Quixote