Machine learning in trading: theory, models, practice and algo-trading - page 953
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
I think we can take a regression channel with a range of 100, shift it on each bar, and if the slope is greater/lower than X, then consider the area described by the channel a flat. What do you think?
(We think almost the same way, but only almost.)) Draw your channel on the chart and you will see.
Do you also convert BP or do you work with what you have, i.e. OPEN, CLOSE on M1, M5, etc., like Asaulenko and others?
How do you know that I only work with OHLC? Although, of course, I mostly work with them, but not only with them.)
But as A_K2, I'm not fooling around.) What they give me, that's what I work with. All kinds of transformations are a loss of information. Either in the process of transformation we get the final result, which does not need further processing and is ready for immediate use, or such transformation is unnecessary.
How do you know that I only work with OHLC? Although, of course, I mostly work with them, but not only with them.)
But as A_K2, I'm not fooling around.) What they give me, that's what I work with. All kinds of transformations are a loss of information. Either in the process of transformation we get the final result, which does not need further processing and is ready for immediate use, or such transformation is unnecessary.
"All conversions are a waste of information" is an incorrect statement.
Transformations can be useful, useless, harmful, and downright stupid.
Transformations can result in different outcomes. In some cases, it is loss of information. In other cases it is an extraction of information that is present implicitly.
"All conversions are losses of information" is an incorrect statement.
Transformations can be useful, useless, harmful, and downright stupid.
The result of transformations can be different outcomes. In some cases, it is loss of information. In other cases, it is the extraction of information present in an implicit way.
This is not an incorrect statement, but certainly true.) I might add, except for the reversible ones.
With the rest, perhaps, you can agree).
We think almost the same, but only almost.) Draw your channel on the chart and you will see for yourself.
The thing is that the variant I suggested is not very good at drawing - there will be 100 channels for 200 bars. You can take the MA channel and look at its slope and width. Another thing is to define the point of exit from the flat, i.e. to formalize when to consider that the flat is over. It is clear that all this will be done retroactively, but it is not important for preparation of the sample.
The thing is that the variant I suggested is not very good at drawing - for 200 bars there will be 100 channels.
There are 100 channels for 200 candlesticks? And this is for determining the flat? That's not very nice.)
There are 100 channels for 200 candles? And this is for determining the flat? That's a pretty big deal.)
The shift will be every bar - the channel window of 100 bars.
The shift will be every bar - the channel window of 100 bars.
So you only need the last point of the channel. What should we do with the rest? That is why we shall draw only the last points.
Instead of the regression line, plot the MA, but not from the beginning of times, but from, say, 100-200 points backwards, or even 20 points if the MA is short. Define it yourself.
What have we got here... more water.
Yuri, at least write something useful (not about how to predict a sine wave with a neural network)
What is there in terms of channel points and MA... the essence only without water.
What have we got here... more water.
Yuri, at least write something useful (not about how to predict a sine wave with a neural network)
What is there in terms of channel points and MA... the essence only without water.
Already said it all. No water).