Machine learning in trading: theory, models, practice and algo-trading - page 3557
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
Few people think about it, but quantisation can increase the number of predictors by focusing on different information from the overall stream.
As an option, here we take deviations from MA (data stream) with equal step, let's say 100, and we get a channel - data on the channel number at different moments of time will tell about the current price position, volatility strength, current trend, trend volatility (for the time of deviation from the average price).
Now let's change the quantisation table and look only at what is happening near the levels.
Such data can already tell us about the situational price behaviour, especially if we assume that the levels are significant for the market, we will be able to evaluate the attempts to break through the level, its resistance to them, and in general - whether the level is significant or not. They can also be signal points for making a decision to open a position.
If the tables are built purely on the scale of indicators for the whole sample, it can be of little use, although the initial data will be valuable, but they will be incorrectly processed.
"Garbage in - rubbish out".
It is pointless to take predictors that are not known to be related to the teacher.
"Garbage in, rubbish out."
It is pointless to take predictors that are not known to be related to the teacher.
Not rubbish will automatically make the input data useful
Useful - working
Working - profitable
If you don't have such - then you don't know whether it is rubbish or not.
And if you have tried it and it doesn't work - that's another matter.
That's the way you should parry. Tried or not and what result (nature of the result)
NOT rubbish will automatically make the input data - useful
Useful - working
Working - profitable
If you don't have such - then you don't know whether it is rubbish or not.
And if you have tried it and it doesn't work - that's another matter.
That's the way you need to parry it. Tried or not and what result (nature of the result)
Well, the phrase "Garbage in - rubbish out" is as good as "That's just it!", suitable for any case in life with equal success, i.e., useless. But it sounds profound.))
"Garbage in, rubbish out."
It is pointless to take predictors that are not known to be related to the teacher.
Few people think about it, but quantisation can increase the number of predictors by focusing on different information from the overall stream.
As an option, here we take deviations from MA (data stream) with equal step, let's say 100, and we get a channel - data on the channel number at different moments of time will tell about the current price position, volatility strength, current trend, trend volatility (for the time of deviation from the average price).
Now let's change the quantisation table, and look only at what is happening near the levels.
Such data can already tell us about the situational price behaviour, especially if we assume that the levels are significant for the market, we will be able to evaluate the attempts to break through the level, its resistance to them, and in general - whether the level is significant or not. They can also be signal points for making a decision to open a position.
If the tables are built purely on the scale of indicators for the whole sample, it can be of little use, although the initial data will be valuable, but they will be incorrectly processed.
You have a fatal error here already at the very beginning. Quantisation implies straight boundaries, not curved ones 😀
Try again
+
Did you erase your last post yourself or did the moderator? Didn't have time to watch the video...
Did you erase your last post yourself or did the moderator? Didn't have time to watch the video...
Original. I'll put it in the topic of what to feed to the neural network input
I thought it was a trivial thing that everyone uses.
This can be done with different channels, fantasise :)
Yes, I thought that it is pointless to discuss this topic further :)
I hope that you realised the fallacy of the statement about the curve of the quantum table, and for this reason deleted.
If I'm not mistaken, you are the one who uses returns from mashka in your articles - I thought it would be clear and understandable.
you need to check the data inside each bin, not their cumulative impact on the target.
You can do it inside - I don't mind - I wrote long ago that this was the original purpose of quantisation - to build models on subsamples of quantum segments (bins).
I just saw additional possibilities that quantisation gives and started to study them in depth.
"Garbage in, rubbish out."
It is pointless to take predictors that are not known to be related to the teacher.
We can only rely on statistics. The process itself is hidden from us - only its consequences are visible.
I used to develop many metrics on this type of predictors, even before I took up MOE.....