Machine learning in trading: theory, models, practice and algo-trading - page 2119
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 read somewhere that ticks are thinning and that all brokerage companies have different amounts of ticks. Some have 100 ticks per minute, some have 300. This is not very often in demo accounts. For example one broker has 1 liquidity and quotes provider, the other one has 3, and the other one combines them.
On something unstable from one broker to another, it is impossible to make something stable.
Yes, exactly! I forgot to add, Demko did some studies on Alpari ticks (I don't know if one can write the broker's name here...).
Stationarity/non-stationarity is not the cause of all traders' troubles, you can improvise a series with the same statistical characteristics as the financial series, the same non-stationary one, but from which it is easy to make a grail. Predicting the price is not the purpose, in order to trade we need to predict the future returnee, the returnee series is quasi-stationary if it is aligned with the seasonal volatility. But the future returnee is predicted very badly, really crappy and it has nothing to do with non-stationarity, what does it have to do with anything, if it were at the cumulative price, it would be an obvious pattern like seasonal volatility, but it is not, so why keep talking about it?
As it goes, the frequencies are floating and maybe the phases... The amplitudes are holding...
Here's a forecast for 500 points of the fitted model on the history of 10k of 4 harmonics
We can see that the forecast is relevant for all 500 points, but frequencies are fluctuating, and fluctuating according to an incomprehensible algorithm
And this is an illustrative example, it may be very annoying
As I go along, the frequencies and possibly the phases are floating... The amplitudes are holding...
Here is the forecast for 500 points of the fitted model on the history of 10k of 4 harmonics
We can see that the forecast is relevant for all 500 points, but the frequencies are drifting, and they are drifting according to an incomprehensible algorithm
And this is an illustrative example, it may be even worse.
So all these "cutting through", without understanding, without adaptivity So all these "cutting", without understanding, without adaptivity to the current situation, are crap!
We need to build an adaptive algorithm that will assess the current situation and adequately change either the series or the forecast.
So all these "cuts" of yours. without understanding, without adaptivity to the text situation, is crap!
On the contrary, I've seen such pictures somewhere in methods of data thinning and reduction of event streams to some heterogeneous time. Thinning in the form of reduction to OPEN prices in equal-tick bars is not a bad thing. But, I won't insist.
On the contrary, I have seen such pictures somewhere in the methods of data thinning and reduction of event streams to some heterogeneous time. Thinning in the form of reduction to OPEN prices in equal-tick bars is not a bad thing. But, I won't insist.
I'm not saying that thinning doesn't work, it is rather necessary, but not in such a primitive form as you are discussing!
There is an adaptive filter that evaluates the current situation and adequately changes its settings.
And there's a stochastic with a period of 14 on the non-stationary market.
When you're discussing ticks and returnees, you're discussing stochastics.
PS Demko by the way is a cool dude, talked to him ...
possible solutions to the synchronization problem
https://stats.stackexchange.com/questions/31666/how-can-i-align-synchronize-two-signals
https://stats.stackexchange.com/questions/16121/for-two-offset-sampled-data-series-what-is-the-best-estimate-of-the-offset-betw/16280#16280
:))))
Interesting comments here:
https://smart-lab.ru/blog/658290.php
I was beginning to think there were really interesting comments, but there...
I was beginning to think there were really interesting comments there, and then...
Well, I don't know... For me, it's the best debate on the applicability of MO to financial series in a long time.
The main conclusion - without finding stationarity, there's nothing for neural networkers to do in the market. And life experience suggests that this is exactly the case.
The simplest example, now that you mention Demko. This is the very person who told me that he has been here on the forum for over 10 years and that this forum is part of his everyday life and he will never leave it.
However, after conducting a series of experiments with OPEN equal-row prices, he, all of a sudden, screamed: "I found what I've been looking for over 10 years!!!! Now you suffer-your turn!" And he left, seemingly forever... I haven't heard from him for 2 years now.
Once again, for understanding - Demko is the biggest fan of forex and there's no need to say that he was disappointed. The only conclusion is that he's found the Holy Grail at the prices of OPEN equal bars, and now he's just stuffing his pockets with cash. So...
Well, I don't know... In my opinion, this is the best debate on the applicability of MO to financial series in a long time.
The main conclusion is that without finding stationarity, there's nothing for neural networkers to do in the market. And life experience suggests that this is exactly the case.
The simplest example, now that you mention Demko. This is the very person who told me that he has been here on the forum for over 10 years and that this forum is part of his everyday life and he will never leave it.
However, after conducting a series of experiments with OPEN equal-row prices, he, all of a sudden, screamed: "I found what I've been looking for over 10 years!!!! Now you suffer-your turn!" And he left, seemingly for good... I haven't heard from him for 2 years now.
Once again, just to clarify - Demko is the biggest fan of forex and there's no need to say that he was disappointed. The only conclusion is that he had found the Grail at the price of OPEN equal bars, and now he's just filling his pockets with cash. Just like that...
we'll all be there... After Alyosha, Asaulenko, Demko, Fokus, Doc... the main thing is not to be in a mass grave
Personally, Demko wrote me some notes in his personal message and, as I understand it, nothing worked for himall of us will be there...
Do you think you followed the path of Alyoshenka the son, whipped by villainous investors? Hmm.... Well - anything can happen...
Do you think you followed the path of Alyoshenka the son, whipped by villainous investors? Hmm.... And what - anything can happen...
It is also necessary to leave gracefully, not with one's tail between one's legs )).
Let's mention Reshetov and his great disappointment in his generator