Bayesian regression - Has anyone made an EA using this algorithm? - page 23

 
Yuri Evseenkov:
What about your works: Fourier price extrapolation, autoregressive (AR) price extrapolation model, linear regression slope, etc. I know that the Dodd-Frank Act prohibits most transactions in the OTC market (including Forex) for U.S. citizens and companies. Let me ask a question. How does the average American buy/sell a little Apple? I watch the TV series "Friends" and there the average masseuse Phoebe not only traded stocks but also became a manager and drained clients' assets....

There is also a predictor based on nearest neighbour somewhere in the base, and a neural network based predictor on 4. These are all early fruits of my amateurish delusion that prices can be predicted by regression or other abstruse models. I have run them quite thoroughly in various runs through history and the result is 50% prediction accuracy. Although the nearest neighbour with certain settings may give 52-53%, but it's not enough to earn robustly. That's why I put all of them for free with the exception of GRNN classifier that classifies pivot points rather well, but with strict filters very few trades are obtained.

 
By the way, shares in Apple and other companies can be bought freely by opening an account with a registered stockbroker. There are a lot of them here. Forex is another matter. It is unknown to many here, and the number of forex brokers registered in the USA is only 2-3. Alpari disappeared, CitiFx handed over their accounts to FCXM, all the others closed down too. In 2013-2014 the US started prosecuting big banks for fraud and forex price manipulation. In 2015, the fall of the franc bankrupted many forex brokers. FCXM and Gain Capital (Forex.com) are probably the only forex firms offering forex trading legally here, but they have not yet moved up to 5. The FCXM, which is considered the biggest forex broker, is now on the verge of bankruptcy and is unlikely to survive for long. I am wary of opening an account with an overseas office that is not registered in the USA.
 
Vladimir:

... I'm wary of opening an account with an overseas office not registered in the US.

Is there a hunt?

Speaking of which, as of late, it has become increasingly common for offices to refuse to accept US citizens.

 
Олег avtomat:
Open books in that direction -- perhaps along with understanding the essence of identification comes understanding of the essence of optimisation.
In a similar situation, I would simply summarise "understanding the essence of optimisation".
 
MikeZv:
In a similar situation, I would simply summarise the 'understanding of the essence of optimisation'.
These are interrelated entities. And a summary is unlikely to give an understanding.
 
СанСаныч Фоменко:


A sample is taken. A subset is randomly selected from this sample and counted as a profit factor on it. Then a random sample is taken again and so on, e.g. 1000 times. You will get 1000 profit factors. This set can already serve as a basis for statistical conclusions.


A subset is selected - are they consecutive values or in random order ?
 
Олег avtomat:
These are interrelated entities. And a short summary is unlikely to give understanding.
I have been taught since childhood by clever people: "If I cannot explain the essence of a theory "on my fingers" even to a child, then I don't understand it myself.
I tried to explain some things from TVIMS to my teenage son, and I was able to do some things, and I am happy about that. So I didn't waste years in vain ... :)
 
MikeZv:
Is a subset selected - are they consecutive values or in random order ?

It varies.

There is when individual observations are randomly selected.

There is when the sample is divided into several parts, e.g. 10. One part with an arbitrary number does not take part in the test, and the rest are tested on. On each of the 9 parts the result is counted and then compared with the one that did not participate. The 9 parts are then randomly selected again and the process is repeated.

The previously described process is a bootstrap.

The one described above is a set of several methods, which differs in how the non-participating part is selected.

 
MikeZv:
I've been taught since childhood by smart people: 'If I can't explain the essence of a theory 'on my fingers' even to a child, then I don't understand it myself.
Some of the things I tried to explain to my teenage son, I managed to do some of them, and I am happy about that. So I didn't waste years in vain ... :)

Keep up the good work.

And the "TC's explanation on the mark" is a hoax. Such an "explanation" may tell youwhat to do, but it won't show you why it's the way it is and not the other way around.

 
Олег avtomat:

Going on the hunt?

Speaking of which, as of late, it has become increasingly common for offices to refuse to accept US citizens.

Is it possible to make a million dollars on forex? If so, who will pay it out? The company through which we trade? Where does it get this kind of money? From clients who pay spreads and lose, that is, by the definition of the Forex market the office loses when its clients win. If all clients win, where does the money come from to pay them all? Insurance? So forex brokers are insurance companies. Clients pay for insurance with spreads and commissions in the hope that these payments will come back to them with 400:1 leverage. But then again, if everyone loses, the broker goes bankrupt. Then back to the original question: is it possible to make a million dollars on forex? Probably not, because it is more profitable for the office to close and appear under another name. But this is a fraud! They have to be prosecuted! Pardon me, what law? The office is opened in a place where the law doesn't exist. How do we prosecute? Do you understand the logic of my thinking?