Machine learning in trading: theory, models, practice and algo-trading - page 2610
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
As conceived, it should search for something - who knows what, but which is logical. This is on the subject of automatically generated strategies. So far I rate it a C, maybe there will be some progress. An untried topic.
I have some scepticism that in principle it is possible to automate the construction of correct predictors from raw quotes. The problem I see is approximately the following. Let's assume that the best predictor is the zigzag links. If we know it, we can build a model that builds a zigzag from quotes. If we don't know it and search for predictors automatically, then there is a high probability that we will miss a zigzag. We can also say that the potential presence of a good model does not mean that we will necessarily find it.
I'm not ready to justify it yet, it just intuitively seems so. For now I try to use predictors and ideas about regularities that are meaningful for technical traders, similar to those described above by my friend Secret. That is to say, for the time being I consider MO not as a basic tool, but as an auxiliary one.
I have some scepticism that in principle it is possible to automate the construction of correct predictors from raw quotes. The problem I see is approximately the following. Let's assume that the best predictor is the zigzag links. If we know it, we can build a model that builds a zigzag from quotes. If we don't know it and search for predictors automatically, then there is a high probability that we will miss a zigzag. We can also say that the potential presence of a good model does not mean that we will necessarily find it.
I'm not ready to justify it yet, it just intuitively seems so. For now I try to use predictors and ideas about regularities that are meaningful for technical traders, similar to those described above by my friend Secret. That is to say, for the time being I consider the MO not as a base tool, but as an auxiliary one.
I do not consider IR as a basic tool.
but some bright ideas are emerging in this matter, i very much hope and therefore i'm reading....
at last ;)
----
my model is elementary to the extreme: action-response.
as they say - the basic building block for everything else
for example, buy - price will go down and vice versa ;)
and I don't consider IO as a basic tool
but some bright ideas are emerging, very hopeful and therefore reading....
finally ;)
----
my model is elementary...
For example, buy - price will go down
Or bought - price will go up...
50/50 - same eggs but from the side...
Or bought - the price will go up...
50/50 - same eggs, but from the side...
if it was 50/50, then I would totally agree
but 99% lose
it turns out that 99/1
you just have to take it for granted and notbother looking for the grail
just have to sit down at the computer and make it;)
---
my wife says sell the software to the bank and i say don't....
If it was 50/50, then I would totally agree.
but it's 99%.
so it's 99/1
;)
Wrong!
In the long term, yes = 99%...
But we are talking about opening a position? ("e.g. bought - price will go down") ... then 50/50
Wrong!
In the long term, yes = 99%...
But we are talking about opening a position? ("e.g. bought - price will go down") ... then 50/50
pay attention please
I wrote everything I wanted to
There's no other way
---
to your post
bought - price down with 0.99 probability
please pay attention
I've already written everything I want to say in there.
There's no other way.
I'm sorry! It's a sloppy handwriting - what I see is what I ... ...I'm not a mind reader...
not Sergei
handwriting is fine
not Sergei
handwriting is fine
About the handwriting...
"bought - price down with probability 0.99"... is that probability 1 or should we add 0.99%...?
About the handwriting...
"bought - price down with probability 0.99"... is that probability 1 or should I add 0.99%...?
99% probability or 0.99% no probability