Machine learning in trading: theory, models, practice and algo-trading - page 797
![MQL5 - Language of trade strategies built-in the MetaTrader 5 client terminal](https://c.mql5.com/i/registerlandings/logo-2.png)
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
......... sharing experience and revealing secrets as it was immediately reluctant. ..........
Share, it is still far away from the secret.
As long as there is a guessing game, you can safely share
By "guessing" I mean determining the order direction to enter the market
Please stop cursing.
Next time, in addition to deleting posts, we'll have to ban you. Do you need it?
Well, this is out of the question. You guys are really screwed now. I wrote such a long post and that's it. What the hell..... The forum glitched... Okay, I'll try to do it again, but it's unlikely to work out so well :-(
Misha, you're grieving, really!
Don't hesitate to write.
First of all I want to turn to you, Maxim, and make an assumption about what happened. Even if I'm wrong, then take it as my situation, because I have been in such situations and more than once, and it seemed to me that with you yesterday something like that happened.
You lost yesterday unexpectedly for yourself, because your previously posted screenshots gave you confidence that at last you have found the method of preparing TS, because the results obtained in your screenshots really looked good. Even I noticed it after a quick glance at them. So you decided to trust the TS and it unexpectedly blew for you. Again, it happened to me and more than once, I just feel comfortable expressing my thoughts to you, but it really looks like it was.
Getting a good result begins to think well, here it is0, finally I found it and the search can be stopped acting on the fine-tuned mechanism. But when you move the interval of training, you get a similar result is no longer possible and again you have to search, modify, change and finally you still get an adequate model, but when you understand that in comparison with the previous training, these changes are very significant just then came to me sincere disappointment, because you understand that you have not found anything, and the last time was only an accident, which occasionally but happens. And in the end each training turns into a mini study where you constantly need to change and look for something to get adequate results. It was this fact that disappointed me the most.
Until the fateful date of March 5, 2018, I also evaluated the model solely on the OOS interval and at the moment when she convinced you, you start to trust her and work on it, but for some reason at that moment she started to fail. I immediately realized that after giving it a work on the OOS it just worked out its potential and when it was time to change it, you just began to work on it. It didn't take me long to understand that the OC interval should be reduced and its adequacy should be determined as quickly as possible, so that there was still time to earn it before it was deflated. But all the same, there was one significant problem, which could spoil all expectations.
You deleted your screenshots of the tests, and I wanted any of them to show the pitfalls you encounter when analyzing the performance of the TS on the CB section. Put some good screencast of training with a section of the OOS and we will analyze the situation and I'll tell you how different the result I got. What is my current advantage..... It will be more obvious...
I deleted posts because no one is interested and no one wants to read books -> nothing to discuss
Regarding plums - over the past year and a half I've earned so much on arbitrage, that it's enough for me for several more years and I'm stuck with the capital-intensive strategy (they don't allow to trade this way on normal deposits)
I have time to make a new capital-intensive TS with emphasis on good investments, or to find a person with TS for investments, including for trading on the stock exchange
I have no idea what to expect from the real broker, but I'll leave it to you to find the right one. I'm writing all my TS from scratch and I don't use any packs, i.e. I understand the whole algorithm from start to finish.
I'm pissed off by all this fuss and unwillingness to get out of my comfort zone and pay attention to other methods.
you just write some nonsense from my point of view, and I don't see any outs for your strategist
Any other questions?
I deleted posts because no one is interested and no one wants to read books -> nothing to discuss
Regarding plums - over the past year and a half I've earned enough on arbitrage to last me for several years and I'm stuck with the capital-intensive strategy (they don't let me trade this way on normal deposits)
I have time to make a new CAPITAL-intensive TS with emphasis on good investments, or to find a person with TS for investments, including for trading on the stock exchange
The real reason for this is the lack of an accurate depiction of the real size of the object. I'm writing all my TS from scratch and don't use any packs, i.e. I understand the whole algorithm from start to finish.
I'm annoyed by this inarticulate fuss and unwillingness to get out of my comfort zone and pay attention to other methods.
Any other questions?
There are no questions. There is an answer. But it sounds like a question. Why? Why look for something when there is a specifically defined algorithm of actions ALWAYS ONE AND ONE that is ALWAYS guaranteed to lead to a good result. I emphasize GUARANTEED. Except for boredom, which is actually why I looked at regression. And if the person does not stop searching and all the time that something searches it only means that it till now so nothing sensible has not found, therefore and searches.... IMHO
There is no question. There is an answer. But it sounds like a question. Why? Why look for something when there is a specific defined algorithm of actions ALWAYS ONE AND ONE that is ALWAYS guaranteed to lead to a good result. I emphasize GUARANTEED. Except for boredom, which is actually why I looked at regression. And if the person does not stop searching and all the time that something searches it only means that it till now so nothing sensible has not found, therefore and searches.... IMHO
In short, I'm interested now in analysis of variance and everything related to probabilities (including conditional) and to work with non-stationary series (new methods, developments)
i have already finished studying mO and took what i needed from it
If I have something to say on the subject I will discuss it, all the rest seems to me nonsense and a step too far.
In short, I am now interested in analysis of variance and everything related to probabilities (including conditional ones) and to work on non-stationary series (new methods, developments)
i have already finished studying mO and took what i needed from it
If there will be something on the subject I will discuss it, all the rest seems to me nonsense and a step already passed.
I see. So you're not interested in MO anymore... Sorry, I didn't know....
I see. So you are no longer interested in the MO... Sorry I didn't know....