Machine learning in trading: theory, models, practice and algo-trading - page 273

 
(I'mnot sure where to get it:

3-5% is an advantage over randomness (53-55%) - 50%, on samples of hundreds of thousands of samples (not chunks).

PS: Dmitry, I see you want to fool around, unfortunately I'm not a good candidate for that. Good luck with white noise and infinite variance.

Not Dmitry, but I'll tell you - not right. White noise is really the easiest thing to work with. Medical fact, sorry. The only question is where to get it, this white noise? ))

 
I'm just trying to get a sense of how itworks:

1) With my own software, of course))

I can tell you something, how to write and what to look for.

2) About the features levels, I will not chew in the mashed potatoes, I've already said a lot,

I think you didn't say anything at all))

If you are a real algotrader,

I am not)), you may say I'm just learning to program

You must have an irresistible motivation to screw it up yourself and a whole bunch of ideas how it can be implemented in your system of abstractions, starting with the lowers, pivot points, etc. You know where you stop, according to what algorithm, these points are "attractors", or rather anti-attractors, from which price bounces, from the top and bottom.

What is this system of abstractions? What are attractors? Is this from the Chaos Theory?

You know where the stops are.

I haven't placed stops after highs and loys for about 5-6 years.

3) Checking the significance is as simple as a bagel. You have a model, (re)teach it with and without this trait, calculate the difference , and this will be the trait's contribution to the model.

I see... nice and simple.

If you don't mind showing me how your model works, just a graph and some signals just to get my bearings

 
Yuriy Asaulenko:

Not Dimitri, but I'll tell you, you're wrong. White noise is really easy to work with. It's a medical fact, I'm sorry. The only question is where to get it, this white noise? ))

How?
 
toxic:

....these points are "attractors," or rather anti-attractors.....

You're on fire...
 
mytarmailS:
An attractor is an attractive point that attracts price.

Lowi hai can both attract and repel, depending on the ratio of limit{stops. Round levels are still a thing to pay attention to. Etc.
 
Combinator:
An attractor is an attractive point that attracts price.

Lowi hai can both attract and repel, depending on the ratio of limits/stops. Round levels are still a thing to pay attention to. Etc.

Are there statistics that justify the existence of "round levels"?

I've been looking for a long time - haven't found....

 
Dimitri:

Are there statistics that justify the existence of "round levels"?



What is interesting, the round levels are set mainly by limits, and stops are set based on price levels of the chart.
It's the same on crypto (in terms of limits, stops are not seen there).
In general, wherever there is a deep glass it can be seen.
 
Combinator:


What's interesting is that on round levels, they mostly put limits, and they put stops based on the price levels of the chart.
In crypto it's the same (in terms of limits, stops are not seen there).
In general, wherever there is a deep glass it can be seen.

If.

This is a picture that shows the positions of players whose bets are NOT placed on the market.

 
Dmitry:

This is a picture that shows the positions of players whose bets are NOT put on the market.

) ok.
 

mytarmailS:

1) I can tell you something, how to write and what to look for

Thank you, for me it is already past the stage, but maybe someone else your advice will help.

2) I am just learning to program.

I cannot do without it.

3) What kind of abstraction system? What kind of attractors? Is this from the chaos theory?

Each programmer over time develops his own programming style, as well as the trader, especially if he is working on his own or in a small company. Therefore, it is much easier to understand an idea and translate it into code yourself, than to pick up the pieces of someone else's code, especially if the other coder has a different way of thinking, which is often the case. The system of abstraction is simply put its own trading API

Mr. Combinator explained about the attractors, you should be able to figure out how to calculate the measure of proximity to the nearest Lions or reversals of the past, this will be the sign.

4) If not hard can you show how your model works, just a chart and signals, just to orientate

I don't think so, I could jinx trades, market observers like to show nice pictures, I don't deal with market observers. If you learn how to trade profitably, you will learn by that time that extra popularity has no sense, it's not showbiz like near-market traders.