Machine learning in trading: theory, models, practice and algo-trading - page 1230
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
Ivan Negreshniy:
overskill and brain hibiscus:)
I'm sorry, I didn't know you were so touchy, well I don't have time to prove to every seller of Expert Advisors and weight loss products that you can easily do regression with scaffolding with boustine and extrapolate, it's not my mission, sorry, go to Gerchik for a start, read what Inokenity offers, I already told above.
Reminds me of
)))), thanks! I haven't laughed like that in a long time!
Maxim Dmitrievsky
I don't think that it would be a good idea to think of the price as a classical time series or as a BP in general
The problem is that with highly liquid symbols the price really includes everything and all strategies, someone has options, someone bought futures (and the spot is linked to the futures), then the time of expiration approached, and then it all gets scattered with a news flow, and then the saggo trade comes out of the shadows)))
Imho, complete chaos is the price on the high-liquid instruments, everything works there but nothing)))
I'm not trying to convince or impose anything, but the guy is making a profit in a 3-4 tick stop in terms of local MO systems or just systems built with functions.)
I'm not trying to convince or impose anything, but the guy makes a profit in trades with a stop of 3-4 ticks, in terms of local MO systems or just systems based on the functional analysis of prices this is the black magic, the impossible, but it exists, I repeat 3-4 ticks stop in the chaos as you say)))
I'm not trying to convince or impose anything, but the guy makes a profit in trades with a stop of 3-4 ticks, in terms of local MO systems or just systems based on the functional analysis of prices this is the black magic, from the category of impossible, but it exists, I repeat 3-4 ticks stop in chaos as you say)))
ok, is there anything on the subject of MO? from the ranks of the possible, please
it's very poor here... either no one is doing anything, just chattering, or from a meager witOkay, do you have anything on the subject of the MO, from the range of possible, please?
Either no one is doing anything, just babbling, or just being poorly clever.No Max, Max, it's just that everyone has tried everything which is generally available to all, and this "all for all" does not work, it is necessary to change the paradigm
No Max, Max, it's just that everyone has tried everything that is publicly available and it's "all for everyone" does not work, we need to change the paradigm
The public is really a nightmarish slag, there's nothing even to read
The public is really full of nightmarish crap, there's nothing left to read.
Max, you don't need to read any more.
The most important thing you already have is the mastery of neural network tools and the Grail on Weierstrass functions. Don't you?
Once again, the market is like Laplace motion.
I attach two artificial Laplace motion series. Try to obtain the Grail either on those series themselves or on the sum of increments (like CUSUM) in a sliding window. If this succeeds, as in the case of Weierstrass, the last step is to select areas (clusters) corresponding to Laplace motion in the real VR according to certain features (kurtosis, asymmetry, etc. for the distribution of increments) and work only with these areas. That's it.
P.S. This post does not mean that I have returned to this forum, full of stupidity, SanSan's grandsons and despair. I just want to help Max and people like him.
Max, you don't need to read any more.
The most important thing you already have is your mastery of the neural network toolkit and the Weierstrass function Grail. Don't you?
Once again, the market is like Laplace motion.
I attach two artificial Laplace motion series. Try to obtain the Grail either on those series themselves, or on the sum of increments (like CUSUM) in a sliding window. If this succeeds, as in the case of Weierstrass, the last step is to select areas (clusters) corresponding to Laplace motion in the real VR according to certain features (kurtosis, asymmetry, etc. for the distribution of increments) and work only with these areas. That's it.
P.S. This post does not mean that I have returned to this forum, full of stupidity, grandsons of SanSan and despair. I just want to help Max and others like him.
Thanks Alexander, I'll put it in the terminal later and have a look.
Max, you don't need to read any more.
The most important thing you already have is your mastery of the neural network toolkit and the Weierstrass function Grail. Don't you?
Once again, the market is like Laplace motion.
I attach two artificial Laplace motion series. Try to obtain the Grail either on those series themselves, or on the sum of increments (like CUSUM) in a sliding window. If this succeeds, as in the case of Weierstrass, the last step is to select areas (clusters) corresponding to Laplace motion in the real VR according to certain features (kurtosis, asymmetry, etc. for the distribution of increments) and work only with these areas. That's it.
P.S. This post does not mean that I have returned to this forum, full of stupidity, grandsons of SanSan and despair. Just want to help Max and others like him.
Another question... can something be taken from this as a substitute? (No Laplace) There is a cool article on BP modeling, maybe you can compose the desired distribution immediately in the terminal
Well, for example - a binomial distribution of returns will do.
Once again - it is very important to make sure that your TS works with such series or not. My probabilistic model works fine with them - another matter is that it is a nontrivial task to isolate the desired parts on real BP, but I seem to have solved this problem.
Working with BP as-built or trying to convert it entirely to the desired form is an unsolvable task, it's not worth wasting time on it. Only work with those parts of BP that you understand and on which your TS works - that's the whole secret.
P.S. And you should not listen to rubbish Asaulenko - he knows a lot and knows nothing. He knows a lot, he knows nothing.