Machine learning in trading: theory, models, practice and algo-trading - page 756
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
Well, do you have the result already? Where is the Expert Advisor?
1) Yes 2) No
1) Yes 2) No
I see. Intermediate results means. You are on the path of intellectual development, which means that every year your system becomes more perfect. Remember the truth: if a person does not have intelligence at 20 years of age, it will never appear. My point is that the maximum period for your development is 20 years. In May 2018 your theory will be 2 years old, so there is still enough time.) Let's wait. I'm just curious how you manage to systematize chaos. They say that it is impossible, but once no one believed in the fact that the earth is round as well, so I wish you success.
I wish you every success.
Thank you.
I'm just curious how you manage to systematize the chaos.
There is an old tried-and-true way, SanSanych has been explaining since the first pages of the thread, but few people listen.
1) Take indicators
2) Analyze whether they are suitable to market forecasting, and delete weak and unnecessary ones.
3) Train the model
4) Test the model taking into account that we have time, and the model must show good results on time intervals, which are unknown to it
Each of these steps is not so simple, but everything is solvable. Much has been written about the second and third steps in this thread, and in general the topic is quite publicly open (data science, machine learning and other similar areas of science).
The fourth step is a little different from the conventional methods. There is an algorithm that gives a prediction, you need to make sure that you can trust this algorithm in the future.
But about the first point - there is nothing good either in this thread or on the Internet. Whoever will learn to create his own indicators from scratch, like Wilder or Williams, will do well and earn lots of money.
So it turns out that anyone who selects settings in the tester, in fact, trains an Expert Advisor. If the tester ran the quotes for 1 year, or even for 3 months, and the result is OK, then why not to believe this prediction?
I think all this learning comes down to backtesting on quotes history and tuning. The Expert Advisor is a trainee- one who selects trading parameters by himself, because this history of quotes and settings is supposed to be continued mentally in time, respectively, every day the Expert Advisor is tested over a period of 3 months - this is more than enough.
And another important thing to know where the advisor is - that is the max-min value of the currency pair divided by 10 parts for example in order to assess the probability of a prolonged fall or rise in price. It is clear, isn't it? If it has started trading in the zone of the min quotes value for 3 months, for example, then it is more likely to buy, and it can be prohibited to sell, for example, and when it goes to the mean values of quotes - then buy and sell according to the indicators.
Forex is much more primitive - people buy it cheap, sell it expensive. The essence of the chaos, it can turn on indicators imitating the correct movement, then reverse and after the advisor adapts, the market will turn around again, well, it happens all the time. That's why there are so many strategies and search in an attempt to curb this Forex. This is all interesting, no doubt, but still, the element of chance will always be, the main task is to have more than 50% of profitable trades.
So it turns out that anyone who selects settings in the tester, in fact, trains an Expert Advisor. If the tester ran the quotes for 1 year, or even for 3 months, and the result is OK, then why not to believe this prediction?
I think all this learning comes down to backtesting on quotes history and tuning. The Expert Advisor is a trainee- one who selects trading parameters by himself, because this history of quotes and settings is supposed to be continued mentally in time, respectively, every day the Expert Advisor is tested over a period of 3 months - this is more than enough.
And another important thing to know where the advisor is - that is the max-min value of the currency pair divided by 10 parts for example in order to assess the probability of a prolonged fall or rise in price. It is clear, isn't it? If it has started trading in the zone of the min quotes value for 3 months, for example, then it is more likely to buy, and it can be prohibited to sell, for example, and when it goes to the mean values of quotes - then buy and sell according to the indicators.
Forex is much more primitive - people buy it cheap, sell it expensive. The essence of the chaos, it can turn on indicators imitating the correct movement, then reverse and after the advisor adapts, the market will turn around again, well, it happens all the time. That's why there are so many strategies and search in an attempt to curb this Forex. It's all interesting, no doubt, but still, the element of chance will always be there, the main task is to have more than 50% of profitable trades.
I do not know about others, but I already have an Expert Advisor, which trades on full automatic. For the past two weeks I have not opened a single trade, and it (the robot) has already completed more than twenty trades and everything is fine. So someone is looking for something and someone has already found it.
And there is no need to put everyone under one roof..... It's Chaos market for you, for me it's quite an understandable movement....
So someone is looking, and someone has already found.
This is Chaos market for you, for me it is quite explainable movement....hilarious)))
hilarious)))
Unfortunately, that's all you can do....
Michael!!! I'm burning!!! Please tell me something about this post. Where to download, I just have to look further I myself....
Michael!!! Burning!!!! tell me anything about this post. Where to download, I just have to look further I myself....
Try looking here
http://neuroshell.noxapredict.com/en_products.htm