Discussion of article "Brute force approach to pattern search" - page 2
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
Everything is optimisation unless you exploit the physics of the market, any system we test first in one area and then in another. I don't use optimisation at all, for that matter. Just fitting it to the site. Here too, just the idea is that this fitting takes place in several stages plus a bunch of filters that keep only the best variants. It's basically an automatic search for patterns in the selected plot. The prettier the variant, the more likely it is to be a variant and not a fluke.
Maybe it will help in some way in terms of analysing and/or developing the idea.
Unfortunately regression is well applicable to physical processes, i.e. to those matters where there are clear patterns or at least a strong enough moment of inertia, in the world of money, there is only the probability that the price will move in some direction - and according to my estimates for simple traders for some strategies this probability is very accurately coincides with 50/50 and it seems to be no problem, but this is on a very large period, and on a short period of 15 black in a row and then "Zero" !!!!
As a result, theoretically it seems that everyone is a millionaire, but in practice the drain after drain....
I agree, the differences from physical processes are significant. Nevertheless, it is impossible to do without regression (autoregression) in econometric analysis of time series.
h ttps:// habr.com/ru/post/267035/
Might help something in terms of analysing and/or developing the idea.
For electricity consumption, the idea seems quite reasonable.
Regarding forex, here is a quote from the comments:
> how does your method work on complex series like usdrub?
> It doesn't. I get a lot of forex traders, I'm tired of them. I don't do forecasting on currency pairs at all.
h ttps:// habr.com/ru/post/267035/
Might help something in terms of analysing and/or developing the idea.
I read it, the idea is interesting but I think it will hardly work on the market. You see, if it were so easy to predict even a candlestick in the future, everyone would already do it ) I had another model, I built SLAU on candlesticks, solved them in the code and no result ). Market models as said above, yes, you can think of something. I have a clear feeling that either you use the physics of the market and write a mat model based on the real physics, taking into account orders, stops and so on and get a small but stable profit, or you just take some model and squeeze all the juice out of it with the help of this bruteforce and hope that the pattern will give you a couple of three times to trade in the plus, and then close the shop and wait for next Friday ) .
For electricity consumption, the idea seems quite reasonable.
Regarding forex, here is a quote from the comments:
> how does your method work on complex series like usdrub?
> It doesn't. I get a lot of forex traders, I'm tired of them. I don't do forecasting on currency pairs at all.
In a properly prepared (processed for it) price series, this method should catch seasonal moments.
i.e. opening/closing of trading sessions, transition of the day, some news. Because the same thing happens time after time.
in a properly prepared (processed for it) price range this method should catch seasonal moments.
i.e. opening/closing of trading sessions, transition of the day, some news. Because the same thing happens time after time.
Good point. By the way, I saw one robot in the market I looked at the chart and was stunned). I downloaded the visualisation and looked at it and it just trades a corridor before closing the day). The thing is that swaps are accrued there and trades become quiet in the absence of pronounced trends ) . There can be many such time points. It is with reference to the server time
Bruteforce is a complete overkill of methods.
But the methods must be appropriate to the nature of the process. And in financial markets, it's a non-stationary process.
(i.e. not only the amplitude but also the frequency of price fluctuations is constantly changing).
But the available analytical methods on financial markets have no mechanism for identification of the 2nd factor (constantly changing frequency).
This requires an identified elementary structure (like an atom in physics, like a molecule in chemistry).
TA figures, Elliott waves and others are not such, as they are identified with a gap in time.
The elementary structure is revealed and its parameters are developed (impulse equilibrium theory).
This structure can be a reference for bruteforce.
Bruteforce is a complete overkill of methods.
But the methods must be appropriate to the nature of the process. And in financial markets, it's a non-stationary process
(i.e. not only the amplitude but also the frequency of price fluctuations is constantly changing).
But the available analytical methods on financial markets have no mechanism for identification of the 2nd factor (constantly changing frequency).
This requires an identified elementary structure (like an atom in physics, like a molecule in chemistry).
TA figures, Elliott waves and others are not such, as they are identified with a time gap.
The elementary structure is revealed and its parameters are developed (impulse equilibrium theory).
This structure can be a reference for bruteforce.
I am not familiar with this theory, but I will have a look at it when I have time for sure
in a properly prepared (processed for it) price range this method should catch seasonal moments.
i.e. opening/closing of trading sessions, transition of the day, some news. Because the same thing happens time after time.
Perhaps it is so, but my small study of daily seasonality was a bit disappointing - if history repeats itself, it does not repeat itself as often as I would like.
That may be the case, but my little research into daily seasonality has been somewhat disheartening - history, if it repeats itself, doesn't repeat itself as often as I'd like.
The diurnal cycle at majors draws an almost stable curve. Its shape is constant and all extrema and inflections occur at the same moments of time. You can (and should) trade on majors using the alarm clock :-)
you must have overdone some research, it happens.