a trading strategy based on Elliott Wave Theory - page 228

 
to Grans
Sergey, I hope for your patience. Explain to me (I may have missed something) why we need a model to describe random residuals and what is "elimination". And it seems to me that "elimination" of random residuals is inherently random. What a wrap-up. :о)


Well... in order to predict the value of the price at the next time Y[j+1], you should add a delta to the current value of the price Y[j]; this delta a*X[j] is a random residue that is modeled:
Y[j+1]=Y[j]+a*X[j].
It would be correct to say "pseudorandom". It is by making an adequate model that fully describes the nature of this residual that we are engaged in.

Flip of a coin, degenerate case of the Markov process X[j]=a*X[j-1]+sigma[j] with coefficient a identically equal to zero a=0 and sigma[j] - a random variable taking on two values 1 and -1. The case of non-arbitrage process - it is impossible to get profit (a=0).

Wiener process, degenerate case of the Markov process X[j]=a*X[j-1]+sigma[j] with coefficient a identically equal to zero a=0 and sigma[j] - a normally distributed random variable. The case of arbitrage-free process - it is impossible to get profit (a=0).

And "elimination" is getting rid of deterministic trends (if any), constant components, cyclic, stochastic...
 
to Rosh

<br / translate="no"> There you go, another thing :) Even a rationale can be given as to why winning after losing has a different probability compared to winning after winning .
"Men don't cry, men get upset" :)


It is possible to give a theoretical justification, but it is better not to your wife, after you have lost everything. Unless, of course, the wife has not studied the course of Mrs. Wentzel E.S. :o)))

to Neutron


Well... to predict the value of the price at the next time Y[j+1], you need to add a delta to the current value of the price Y[j], this delta is a*X[j] and is the random residual that we model:
Y[j+1]=Y[j]+a*X[j].
It would be correct to say "pseudorandom". It is the compilation of an adequate model that fully describes the nature of this residual that we are concerned with.


Sergey, I probably have not expressed my thought very correctly. But it seems to me that random residue is obtained (in general case) after subtraction from primary data, some analytical function. This could be LR, superposition of several basic harmonics, etc. This approach to prediction will be correct if this basic function remains "true" at the next iteration. Will it stay? I've already complained that all my attempts at such prediction failed. In general, it turned out like with a coin.
 
Sergey, I can say one thing: if the initial series contains a trend or harmonic component, then AR model does not apply in principle, which means that Markov chain is not applied to describe it either. In this case, other models designed to identify and adequately describe deterministic trends or harmonic components are used. Thus, our task boils down to:

1. Formulation of selection criteria for a particular model of the market.
2. Selecting a model.
3. Construction of an adequate forecast model.
4. Predicting the price in order to make a profit.
 
It makes perfect sense. Where can I read about AR models? Because my knowledge of them begins and ends only with signal processing, and that is very superficial.
 
It makes perfect sense. Where can I read about AR models? Because my knowledge about them begins and ends only with signal processing and that is very superficial. <br / translate="no">.

To be honest, I have bits and pieces of information on this topic. Maybe Northwind can share his materials, if he has any...

Here's what I was thinking: I propose to hang this picture from my dissertation in my office in the most prominent place for every sensible trader.



In fact, this is a demonstration of the soundness of the theoretical model proposed by Pastukhov. Now every now and then you hear talks about the fundamental impossibility of reliable earnings in the Forex market. The result is intended to increase confidence in our ranks in the rightness of the chosen path and the desire to continue what we started.
There is, however, one subtle point... Imagine that you are a graduate student and you have received instructions from your supervisor to calculate the parameters of the model describing the evolution of the system of two gravitationally-bound stars. Let's suppose, it was lazy for you to study Einstein's General Theory of Relativity and you have made "wise" decision to deduce all formulas from the general considerations independently - it is after all interesting and instructive. Undoubtedly, we will succeed and we will triumphantly put on the table already the son of the former leader the sought model...
Maybe we should throw away all our bicycles and pay close attention to the study of what has already been written in the thesis. Maybe it makes sense to do it together? I'm sure it will save a lot of time and WARRANTY lead to a positive result - material well-being.

Who's counting?
 
...Who thinks so?

I think so, but I have absolutely no free time at the moment.
 
...Кто как считает?

I think so, but I have absolutely no free time at the moment.


If someone showed me the bike... I'd study the blueprints carefully and maybe I wouldn't have to reinvent it. Although it's not a fact, one becomes a wise man after a pile of invented bikes, but no one can avoid this path anyway (I think so).
 
...Who thinks so?


On the whole, I support that it's worth looking into this material more closely. The only thing is that I won't be ditching my bike. Especially as it already resembles a "Harley-Davidson". Sergey, the road you walk is very important. If you walk on a beaten track, it's possible that plantains don't even grow on it. I think it's quite clear what I'm alluding to.

I can only add that it's also a bicycle.
 
Rosh 23.01.07 10:52
If someone had shown me that very bike... I would have studied the blueprints carefully and maybe I wouldn't have reinvented it. Not a fact though, one becomes a wise man after a pile of invented bicycles, but no one avoids that path anyway (I think).

Of course the bike in question is the one invented by Pastukhov (reference to the blueprints was given).
 
Colleagues, sorry for the flood, but I couldn't resist. A smile and a good mood shouldn't leave us, after all.

"After the auction."
http://grasn.narod.ru/img/t.JPG

PS The jpg display is not working
Reason: