Demonstrating the cluster approach to market... - page 2

 
sol >> :
And what is the voltage around August 2008 and where are the currencies discharged by this indicator?

More specifically, on the night of 7 to 8, around 3 o'clock...

 
ssd >> :

What we have here is that we are adding / subtracting points whose "weight" may vary.

clusters should be made of oscillatory indices, i.e. without a constant component

In oscillators, the difference (A-B) should be replaced with the ratio (A/B-1)

points were invented for the convenience of the human mind to perceive numbers

 
ssd >> :

In order to study cluster indicators I wrote a program


here's something like this

Files:
 
KONDOR >> :

like this too.

This is all understandable. How is the very idea of a "pull" of the currency price towards the zero line?

Is such a "pull" detectable in practice ?

It so happens, that everybody is trying to "improve" the S.S. indicators for himself, but I don't think so.

By testing this key hypothesis empirically, so to speak.


I haven't found that much traction so far. Rather the opposite, as long as, say, an hour / hour and a half, the currency is overbought / oversold.

If the currency price moves into an overbought/oversold condition, then it most likely continues its direction of movement towards the upside

overbought/oversold.

 
sab1uk >> :

clusters should be made of oscillatory indices, i.e. without a constant component

In oscillators, the difference (A-B) should be replaced by the ratio (A/B-1)

points were invented for the convenience of the human mind to perceive numbers

Good point, thank you.


This is where I corrected it:

 

Semenych's idea itself is excellent. Naturally, much that follows from it is not developed to its logical conclusion and needs to be understood independently.

равновесным и естественным состоянием для валюты является состояние вблизи нулевой линии,

I believe that the "dynamic equilibrium" is not the zero line, but the "maximum entropy" state of the currency (let's call it A). The entropy function makes more sense, in my opinion, if we are talking specifically about equilibrium.

Roughly speaking, if we consider a "gas" consisting of "molecules", each of which is a pair containing currency A, and encode the states of each molecule as -1, 0, +1 (the movement down, in place and up of currency A), then, having made up a "microstate" of the whole system, find its equivalent and renumber them, we will get an expression different from the one proposed by Semenych.

We can talk about it for a long time, but a few consequences of such classification can be mentioned. Let's assume that our gas contains 9 molecules (i.e. 9 pairs containing currency A).

1. The most probable state is the "gaseous flate" in the microstate -1, -1, -1, -1, 0, 0, 0, +1, +1, +1, +1 or any other equivalent state. In other words, it's three pairs up, three pairs down and three in place. Any three. This is the most common wide flat that pipsmers love so much.

2. The dead calm, often seen before strong news, i.e. all zeros, is a "crystalline flat". This flat is qualitatively different from a normal high entropy flat - like heaven from earth. In terms of state probability it is equivalent to a powerful trend (all -1 or all +1). There is a hypothesis that it is equivalent to a powerful trend not only in entropy, but so far I cannot substantiate it. However, it is well known that such a "flat" is extremely explosive and can instantly turn into a powerful trend.

3. It is quite possible that it would be interesting to investigate the dynamics of transition of microstates into each other, but I have not done it. Maybe, there will appear some other invariants and regularities, not visible from the point of view of entropy. If I am not mistaken, in works of classical theoreticians of statistical physics there should be something similar.

4. I've been trying to find the usual trade equivalent to the notion of state entropy - and I think I've found it. It's liquidity: the higher the entropy of state, the higher the liquidity.

5. One last thing. If there are any experts in statothermodynamics here, we could look for other thermodynamic potentials (like enthalpy, free energy, etc.) of our gas.

And any degree of "overbought" and "oversold" is a "stressed" state for the currency, from

which it tends to exit by "discharging" towards the zero balance line.

It is rather a matter of time. Before such a "tense" (low-entropic) state "defuses" to a wide flat (high-entropic) state, a lot of water can leak out. So it is not known where to enter correctly. Research is needed here, but the topic is extremely benign.

 
KONDOR >> :

The market takes everything into account. there are no overbought and oversold currencies.

And if any imbalance were to occur, it would be immediately compensated for.

Bill Williams Trading Chaos

Bill Williams, of course, is a master of childish hoaxes.

However, in my opinion, if we look at his work without the "classic" halo that everyone is accustomed to,

he didn't really do anything useful for the practical trade.

"Five bullets that kill the trend....", "Alligator, jaws, lips...." and the rest in the same vein ...

And we've been wondering - "Is the market currently flat ? or Trending ?" and still are...


Pardon the lyrical digression. I really don't like this B.W.

 
Mathemat >> :

Semenych's idea itself is great. Naturally, a lot arising from it is not developed to its logical conclusion and needs to be understood independently.


Wow, what a passage you have!

I will read it all carefully and think about it.....

 

I did some research on liquidity:

I collected tick data, with total liquidity of several interbank venues. I was interested in how the total liquidity for buying and selling changes as the price changes. The results have not yet figured out how to apply in practice, but the direction of research so far seems promising. If anyone needs it, I will post the data on the majors in a format that is easy to import into MathCad and other mathematical packages.

 
Mathemat >> :

yes, liquidity is something i often think about too

and the more people think about liquidity the more people will ask metaquotes for tick history

and the more people will understand that a proper broker should have a floating spread and that 4 digit quotes are sabacha shit

Reason: