Machine learning in trading: theory, models, practice and algo-trading - page 1419
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
With all due respect (you know...), but, perhaps, I will support Asaulenko. Almost one and a half thousand pages, and the conclusion was born one: it is necessary to look for predictors.
Where do you get the conclusion? Well, it's very simple: people place candles in MO and get 50/50. If we attach indicators, the error will decrease. But, gentlemen, if you put a hypothetical Grail that always guesses, you will get 0 errors in the output of the MO. If you have selected a grail set of indicators, you don't need MO. You can do with simpler means.
simple and straightforward....
I have nothing to say to you because there is no issue to discuss. Just remember: the surest way to be deceived is to think you are smarter than others. (с)
Yes, my NS system has been working for more than a year, without any retraining. The very first, still trial, trades here in this thread. New system? - I'm in no hurry.
You are pathetic, you are miserable.
I offered to discuss the topic, to think together on the solution of an interesting problem - zero constructive suggestions and only one person reacted at all to the proposal.
But to sling mud at each other - it is quickly possible, here it is not necessary to strain the brains...
I offered to discuss the topic, to think together on the solution of an interesting problem - zero constructive suggestions and only one person reacted at all to the proposal.
And here to throw slurs at each other - it is quickly possible, here it is not necessary to strain the brains...
Here you can not discuss any ideas, even as simple as GEPs, as here inhabited by reindeer who are very bored with it, because they have long learned the Tao. So you're on your own. Or here's about brain strain
I offered to discuss the topic, to think together on the solution of an interesting problem - zero constructive suggestions and only one person reacted at all to the proposal.
But to sling mud at each other - it is quickly possible, here it is not necessary to strain the brains...
I have an alternative proposal for unions. All who suggest it should unite with each other and show us all the miracle of joint action) For example, such proposals were in a thread TA. Unite, unite with each other at last!)
I propose to work together on predictors based on the phenomenon of gaps.
On the screenshot you can see that gaps are often subsequent correction points - both large and small - I think that for those who like to work on small movements it can be useful.
Very interesting topic... How do you see gaps as predictors ? Because the conventional form of predictor is a vector (column) with values in a matrix, and in the case of a gap there are many parameters:
1) the fact that hepatitis is present
2) the hep price
3) Absolutely different time of action of the gel, one gel will work in 15 minutes, and the second in 4 weeks, it is not clear how to create it as a dataset
4) The total number of geops, it is also not clear how to create a dataset.
How do you look at these problems, maybe there are some solutions?
I tried to look for such points that the price almost always comes back to. I found quite a few "patterns" in the market, the price returns to them with 90% probability, and the chart looks very similar to yours, the price not only comes to the expected price, but often changes its direction when it reaches the target
You can't discuss any ideas here, even simple ones like GEPs, because there are reindeer here who are very bored with it, because they learned Dao a long time ago. So you're on your own. Or here's one about straining your brain.
Honestly watched the lecture again, about the lecturer's delusion about the knowledge of psychology will not - manipulation of historical facts to increase sales, talked about this earlier. Overall, a good quality mechanic with a deep knowledge of the structure of the brain, which makes it interesting information about the subject from his mouth for the average person. Another thing that surprised me was his unproven hypothesis about the origin of the frontal lobes and the whole new brain structure, without discussing the hypothesis itself, but only drawing attention to the attitude to it as a key to the mystery of the origin of the new layer of the cerebral cortex, whose arguments lie in zoopsychology. And, take such social animals as elephants, where the care of offspring is no less long than in primates - 10-12 years - why is there not such a pronounced sign of new cortex rudiments as in chimpanzees? In general - it would be interesting to look at the scientific evidence, which apparently is lacking in this theory.
I have an alternative suggestion about associations. All who suggest it should unite with each other and show us all the miracle of joint activity) For example, such proposals were in a branch of TP. Unificationists, unite with each other at last!)
Why do you have to show something - do not work for the same budget :)
I would work on a base of predictors - I have some experience and methodology, but I would like reciprocity, otherwise what for do I need it.
Very interesting topic... How do you see geps as predictors? After all, the generally accepted form of predictor is a vector (column) with values in a matrix, and in the case of hep there are many parameters:
1) the fact that hepatitis is present
2) the hep price
3) Absolutely different time of action of the gel, one gel will work in 15 minutes, and the second in 4 weeks, it is not clear how to create it as a dataset
4) the unlimited number of geops, it is also not clear how to create it as a dataset
How do you look at these problems, maybe there are some solutions?
I tried to look for some points, to which the price almost always comes back. I found a few "patterns" in the market, the price returns to them with 90% probability, and the chart looks very similar to yours, the price not only comes to the expected price, but often changes its direction when it reaches the target
I often use several columns to describe the same phenomenon - the problem is how to teach the model to work with a group of columns reflecting the same idea. Predictors are like constructor parts in this case that fit together within the same group.
I personally see the description of a gap as creation of a sort of a cup with gap levels, which will contain all necessary information with a limited depth - for example 5 gaps above and below the current price. Accordingly, the characteristics of each gap will be given - the gap, the distance to the gap, the number of bars passed since the gap occurred - this is the minimum that can be expressed as natural and relative indices.
A similar method can be applied to other levels.
I often use several columns to describe the same phenomenon - the problem here is another, how to teach the model to work with a group of columns reflecting the same idea. Predictors are like constructor parts in this case that fit together within the same group.
I personally see the description of a gap as creation of a sort of a cup with gap levels, which will contain all necessary information with a limited depth - for example 5 gaps above and below the current price. Accordingly, the characteristics of each gap will be given - the gap, the distance to the gap, the number of bars passed since the gap occurred - this is the minimum that can be expressed as natural and relative indices.
A similar method can be applied to other levels.