Machine learning in trading: theory, models, practice and algo-trading - page 1644
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 misspoke. To find the "optimal points" for entry/exit (on the history) you can But this requires defining the criteria of a "good deal" and writing a special algorithm that analyzes the history in the context of dynamics. It will select suitable trade areas and optimal entry/exit points for deals. Then, on these points we will test the parameters of the indicators, and if their values in the points are repeated, it means they really "follow" the market and can bring profit. Then we include these parameters in the TS signal.
In my opinion, we all want to make a profit, we want the indicator to work as well as in hindsight, such as ZZ, but how to do it?
In fact, you suggested choosing and fitting the indicator parameters to the history and optimizing their values on the fly. In that topic, the task is similar - you need to automatically select the successful parameters (from the prepared sample) for the signal, checking their values in the "ideal points" on the history. If the values in the parameters are repeated at the points, it means that they are good for the TS.
The problem: the criteria of "ideal points" were not found. Therefore - it was not possible to find these points on the history and therefore - to select "adequate" parameters for the TS signal.
Deadlock.
Well... not the right way to look, maybe? ;)
Did you use indicators?
More likely, yes.
What kind?
Stochastics and stuff... (the most primitive filters) that no one who knows anything about them ever uses.
Are the indicators have fixed parameters?
Yeah...
Is the market stationary or is it constantly changing?
It changes...
Do you take this into account?
No...
And the market is a complex mutual layered process or a regular "two-dimensional"
I guess it must be complex...
How do you account for this complexity?
you don't...
..................
........
...
sorry, I got carried away ...
In my opinion, we all want to make a profit, we want the indicator to work as well as in hindsight, such as ZZ, but how to do it?
ZZ is from my point of view a completely useless indicator.
The history itself is the best indicator possible.
(The concept is simple).
Interesting
BitForex does not allow "algotrading and scalping". Now I use them as a source of correct quotes.
Bummer (!)
BitForex does not allow "algotrading and scalping". Now I use them as a source of correct quotes.
strange requirement, as for a real exchange, but reasonable for all sorts of kitchen scams, when the purpose - the client's deposit, probably will also go the way of BTC-e, or some other similar scam
Is the market stationary or is it constantly changing?
It is changing...
No one is arguing with that. I only believe that the "amateur radio" methods you propose are not suitable for solving the problem of nonstationary prices. The main reasons, in my opinion, are the following:
1) Prices are NOT the output of some linear system - their behavior is much more complex.
2) There is NO "correct" way to divide prices into useful signal and noise - any such division is arbitrary.
No one is arguing with that. I only believe that the "amateur radio" methods you propose are not suitable for solving the problem of unsteady prices. The main reasons, in my opinion, are the following:
1) Prices are NOT the output of some linear system - their behavior is much more complex.
2) There is NO "correct" way to divide prices into useful signal and noise - any such division is arbitrary.
Unfortunately, few people here understand this and agree with it.
2) There is no "correct" way to divide the price into useful signal and noise - any such division is arbitrary.
That's how the unknown Nikolaev took and buried the whole mathematical statics along with econometrics...
That's how the unknown Nikolaev took and buried the whole mathematical statics along with econometrics...
Uh... He recently buried Dialectics, along with the Academy of Sciences commenting on it).
Unfortunately, few people here understand this and agree with it.
To be fair, some consensus on the inapplicability of the Fourier decomposition/fourier transform to prices is apparent here. From here it is not far to understand the inapplicability in general "amateur radio" approach to prices.