Machine learning in trading: theory, models, practice and algo-trading - page 2035
You are missing trading opportunities:
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
Registration
Log in
You agree to website policy and terms of use
If you do not have an account, please register
Maximka go ahead and drag
If my working systems don't work, I'll have to go into artificial intelligence, if you're the best there, I'm luckytry the factory
Several times remade, after unpacking will take 6GB.
Day of week, day of month, hour, minute, ...same for exit..., deal duration in minutes, SL, TP, result +-1I can't make pictures, but I can pull data from predictors, but I'm not too good at the question of type of useful data, here are possible options and choose the one you are interested in
Possible values:I've been tweaking the optimizer lately, mostly in the area of metrics. I have made such a mess that it makes me proud. I am such a great optimizer, and I can play a nice tune in general, so here I am :-)
So, tell us about your achievements substantially ...
OK, yes.
not much, but it is the most plausible description of an abstract TS or more accurately fits the formulation of the problem of searching for TS
I would still like to put emphasis on the inconsistency between the concept of algorithm and TC. Rather, the TS is not a fixed code, but the process of changing it) "Put the Expert Advisor on the chart and forget about it" - this is, in my opinion, an unattainable ideal)
- How long is the life cycle of the TS? (Trader's folklore about testing over 10 years and 10 nights, "sucked" out of the trader's finger, that turns everybody's heads - we should not take it into account)
Testing, in my opinion, does not allow us to find a good TS, it only allows us to discard the bad ones. If there was a very profitable strategy in the recent past, it (or similar ones) will surely be found by other players, which according to the game model will sooner or later lead to its unprofitability in the future.
- what are the tasks of optimization/reconfiguration ?
For myself, I formulate it as "with luck get more than lost with bad luck". When formalizing this principle, it is usually something like maximization of expected payoff considering spread, drawdown limitation and finiteness of the TS lifetime. Of course, all this is done within the framework of some statistical models that the market does not have to strictly comply with.)
No, I haven't) what will be the result? Most likely, I think the results will be very different.
The result will be the same, with small deviations (there are many reasons for that, one of them is rounding) and that's not the most critical, the most critical is the time it will take to get this result and it will be many times larger, if you use MQL as an engine, I spent a whole year on it and in the end I got back to the old and native C++. I'm not trying to discourage those who want to do it, it's a great experience but I don't see why you need it)))
Try the factory.
your ass, I like it
Your ass, I'll second that, I like it.
The result will be the same, with small deviations (the reasons are many, one of them is rounding) and that's not the most critical, the most critical is the time it will take to get this result and it will be many times more, if you will use MQL as an engine, I spent a whole year on it and finally came back to the old and native C++. I'm not trying to discourage those who want to do it, it's a good experience but I don't see why you need it)))
Experience is important)))
I can't make pictures, but I can pull data from predictors, but I'm not too good at the question of usable data type, so look at the options and choose the one you're interested in
Possible values:I don't know, let's go with the first one