Bayesian regression - Has anyone made an EA using this algorithm? - page 7
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
What is the point? Of course you can find it if you look, since the markets have come up with all sorts of different indicators, each of which can be attached to something with greater or lesser success.
СанСаныч Фоменко:
The Party teaches that applying linear regression to price series is incorrect because they are non-stationary. :)
The author in the above post just misspoke. Statistical packages have standard procedures for processing time series: highlighting the trend, highlighting the seasonal component and taking the difference. The author meant the first one.
It's not just one place, there are several times in the whole section. Did a little reading. The article is in the worst traditions of the most dull student textbooks.
Another thing I found: "All Box-Cox transform examples given earlier refer to the case when the distribution law of the original sequence is assumed to be normal. ".
How on earth did you give a link to this article, aren't you an adherent of the "anti-normal distribution" movement? Or are you not?
How on earth did you link to this article, you are an adherent of the "anti-normal distribution" movement? Or are you?
When we try to apply statistics, the cornerstone, the foundation, is the question of the APPLICABILITY of a particular tool from that science.
Your example contains no random variables - a constant. Dispersion refers ONLY to random variables. In your particular case, there was a result unique to statistics: the variance calculation showed that constants, not random numbers, were supplied as input.
The uniqueness of your example is that the result is correct and easily explainable. Usually, if you do not carefully justify the possibility of using a tool, such as linear regression, a result will be obtained that has nothing to do with reality, and therefore completely unusable in practice: numbers will be, they can be seen (gopher visible), but in reality, all these numbers do not! Just a numbers game.
Using linear regression as an example: a standard algorithm (not a homemade one) calculates the regression coefficients and, usually, the far right column tells us whether the regression coefficients we see actually exist. If the far right column has a figure of 0.5 (50%) then it is certain that the printed figures do not exist. If it's 10%, then it's just so, in the fog. but if it's less than 5%, then the numbers really exist. And this can only be believed if you have managed to justify the POSSIBILITY of applying this very linear regression beforehand.
And what makes you think they are not random numbers. Random numbers suddenly can't fit into a straight line? They can also draw a Repin painting.
And in general, the conversation was not about the data, but about the formula, what it represents and what meaning it carries.
How on earth did you link to this article, you are an adherent of the "anti-normal distribution" movement? Or are you?
In fact, the conversation was not about the data, but about the formula, what it represents and what meaning it carries.
As if you've fallen from the moon. As if identifying waves is complicated. The main problem of tehanalysis and consequently of trading is the identification of the trend.
Although the question is not for me. Allow me to introduce myself: A soldier in General Gauss's army. Private, untrained.
In the first post of the author of this thread there is a description with formulas in pdf format. I wish I could find an adequate translation. https://www.mql5.com/go?link=https://arxiv.org/pdf/1410.1231.pdf
Agreed. If one could know exactly whether the market is flat or trending, most high frequency experts would be profitable.
In this post, section 5, trend taking
https://www.mql5.com/ru/articles/363
the author shows a perfectly acceptable approximation of the sample of increments to normal. It has long been known how to deal with points which do not lie on a straight line - they are excluded from the sample by about 7-10% of maximum modulo values. Then even Kolmogorov's goodness-of-fit criterion (which is very sensitive to the form of distribution) shows that the sample is normal. As for the excluded values, these are the points where the current trend has broken down. The source where this methodology came from (I read something in English a long time ago, I don't remember where) basically suggests forming samples of increments from points that are between the trend break points, this is what is suggested to be called the current trend.
This is interesting stuff.
One important note: the author writes that for linear regression and ANOVA a normal distribution of the data is assumed. This is a very lengthy and incorrect statement that many people repeat without thinking. It is, in fact, about assuming a normal distribution of model errors. The data itself may not be normal.
Bayesian regression, linear regression, neural networks, evolutionary algorithms..... eh how rich is the market sucker community.... and how happy professionals are that there are fools who believe in theirscientific models............)
how amazing that it is still not clear that the market is a simple thing... complex algorithms -- screw up because they are just not relevant...
but no - go ahead, so much the better for those who don't jack off on maths, but rather draw resistance levels, watch for false breaks, build up a position and ..... the rest is unknown to most of the forum (it's getting banknotes at the ATM)
we fly and you crawl fools you fools...........
Though the question is not for me. Allow me to introduce myself: A soldier in General Gauss's army. Private, untrained.
In the first post of the author of the thread there is a description with formulas in pdf format. I wish I could find an adequate translation. https://www.mql5.com/go?link=https://arxiv.org/pdf/1410.1231.pdf
Agreed. If one could know exactly whether the market is flat or trending, most high frequency experts would be profitable.
...