The "Maybe we'll get lucky" counsellor - page 9

 
C-4:


Memory is the dependence of the future on the past. Today's change in prices takes into account yesterday's state. The pattern is as follows:

Time
Event

Actual

Response

Implied

EMH response

t1
a1
0
3
t2
a2
0
2
t3
a3
0
5
t4 a4122

I.e. we can see that event a4 "remembered" events a1-a3 which were not played back, and played back the reaction for them at the moment t4. Hence such a clear clustering of price changes, which by the way the same EMH can explain very sloppily and implausibly.

This is not just my opinion. You are not questioning my conclusions, but the very serious meaningful work of people who understand what they are writing about, and this is serious competition.

Once again, time dependence does not follow from having a memory. Lack of dependence on time does not mean lack of memory. Lack of time dependence means that time is not an independent variable. There are many processes with memory that do not depend on time.

For example, plastic deformations, although occurring over time, are dependent on the loads that acted on the body. Elastic deformation works within some limits: memory - when a load is removed, the system returns to its original state, when applied, to the state it was in at the same load some time ago. If a load exceeds a certain limit, the system can undergo plastic deformation and find a new position of equilibrium, and its properties will change slightly - and so on until failure. The state of the system is not dependent on time - only on the history of applied loads. Only in the case of cyclic (periodic, stationary) loading within the elastic limit, the time dependence of the state of the system can be derived indirectly. In other words, time dependence is more a characteristic property of the mathematical model, which describes the state of the system, than the system itself.

 
VladislavVG:

Your absolute confidence is encouraging - perhaps everyone else doesn't believe you simply because they don't know how to use the tool provided. Me too, by the way - here is an example, taken at random from history - your indicator readings with the same hindsight and immediately on history, so you don't have to wait several years to analyze signals..... can you formulate decision-making rules ? :

The rules you suggested - by coincidence of all 3 lines - marked with arrows : no advantages with respect to muvings or, for example, smoothed Haykin were found.

ZS and by the way, Yusufhoja, about your 18 : you will be surprised, but the price is not time dependent. More accurately, price changes are not dependent on the passage of time, although they do occur over time, but they are dependent on other causes and changes are not the same thing. Therefore your article, after such assumptions:

You do not need to read it further.

Vladislav, I agree with your arguments partially. The indicator, which you cited, is an uncorrected variant, and its correction, as I have previously noted https://forum.mql4.com/ru/38834/page270, is very simple:

On exel I made this fix with a single X column:

d=ЕСЛИ(X2=0;ЕСЛИ(F3+G3+H3<0;ЕСЛИ(H3=G3;V3-W3;0);ЕСЛИ(H3=F3;0;V3-W3));X2)

Columns F, G and H here take values +1, (buy) or -1 (sell) depending on the market condition and algorithm calculation.

Columns V and W are calculated values of the red (buy) and green (sell) lines.

d - the appropriate addition to the former calculated (blue) line of the trader.

That is the correction.
Now the P3 trader line does not make a jump at the moment of anticipated or completed price reversal, but immediately shows the current state of the market. Moreover, it describes historical data perfectly, but, as noted above, contrary to common sense, it is unable to make adequate forecasts beyond the current bar or it is impossible in principle, because we are not destined to know the future, there is a taboo, which even (18) cannot overcome! On the other hand, where have we learned to predict the future - nowhere! Why do we think it is possible in forex? Therefore (18) reminds us - history can be described, but only depending on the past tense! That's why I agree with you partly on the issue of price versus time. The price depends strongly on the past time and this means that we do not have the right, or rather, the possibility, to judge the time so globally, because we are not fated to know the future, not only events, but the time as well! And some successfully made "predictions" are nothing but a coincidence. Now the question is what to do? It is not necessary to do anything. One must only speculate, not predict.

 
VladislavVG:

Once again, it does not follow from the presence of memory that there is a dependence on time. The absence of time dependence does not imply the absence of memory. Lack of time dependence means that time is not an independent variable. There are many processes with memory that do not depend on time.

For example, plastic deformations, although occurring over time, are dependent on the loads that acted on the body. Elastic deformation works within some limits: memory - when a load is removed, the system returns to its original state, when applied, to the state it was in at the same load some time ago. If a load exceeds a certain limit, the system can undergo plastic deformation and find a new position of equilibrium, and its properties will change slightly - and so on until failure. The state of the system is not dependent on time - only on the history of applied loads. Only in the case of cyclic (periodic, stationary) loading within the elastic limit, the time dependence of the state of the system can be derived indirectly. In other words, in this case, time dependence is more a particular case of the property of the mathematical model describing the state of the system than of the system itself.

What is the cause of memory in relation to the market?

- The unconditional reflex of the crowd, generating "price disturbances" incomprehensible to their creators.

- Economic axioms, laws.

- Physical...

 
yosuf: It describes the historical data perfectly, but, as noted above, contrary to common sense, is incapable of adequate forecasting beyond the current bar or it is impossible in principle, because we are not meant to know the future, there is a taboo that even (18) cannot overcome! On the other hand, where have we learned to predict the future - nowhere! Why do we think it is possible in forex? So (18) reminds us - history can be described, but only depending on the past tense! That's why I agree with you partly on the issue of price versus time. The price depends strongly on the past time and this means that we do not have the right, or rather, the possibility, to judge the time so globally, because we are not fated to know the future, not only events, but the time as well! And some successfully made "predictions" are nothing but a coincidence. Now the question is what to do? It is not necessary to do anything. One should only speculate, not predict.
Yusuf, why don't you write books about brutal forex and (18), which is a gigantic powerhouse crashing down... etc.?
 
Mathemat:
Yusuf, why don't you write books about brutal forex and (18), which is a giant power collapsing... etc.?


Not everything at once, Alexei, especially when the Rogun hydropower plant is not completed and everything else, books with memoirs later, it would be better to deal with fractals here...

One gets the impression that the front page will be left with the topics of Treugs, who are constantly on the offensive, villagers with ILANO-LOCKS, and the threads of Yusuf's regular discoveries and transitions to the next spiral...THAT is how it should be + EURO-FLUD, of course.

Yusuf, keep it up. Prepare a thread with discussion of the next indicator for the general case at n=n, how does the programmer feel, by the way??? testing the indicator code or not???

 
yosuf:

The price is strictly dependent on the past tense and it turns out that we have no right, or rather, no possibility, to judge the time so globally, because we are not destined to know the future not only of events, but also of time! And some successfully made "predictions" are nothing but a coincidence. Now the question is what to do? It is not necessary to do anything. One should only speculate, not predict.


Once again, one last thing: the current price does not depend on the passage of time. It depends on actions the traders took, for various reasons: which orders were placed, in what sequence, and on actions the matcher took: in what sequence the orders were satisfied. The time is just a counter for readability - nothing more. No matter what you have corrected there: this model - the model describing the dependence of prices on time - does not fit the process - is that clearer?

This doesn't just apply to your 18 - in general any attempt to "pull" the time-dependence of price on the market. Indirectly that time is involved as no one will wait indefinitely for profits, losses or "sitting on the fence". Everything else is simply a fitting of an inappropriate model to the process.

 
VladislavVG:

There are plenty of memory processes that are not time-dependent.

All very nice and convincing, but only when building models of closed systems, i.e. for dead things. Here mathematics will always be at its best. But the market is a live and open system, capable of accumulating and synthesizing information, and hence, self-organizing. And the price has a memory and time cycles are also quite tight. It is especially evident before and after the news release. Then the news release comes out, but the price does not care about it. Or vice versa, the news has come out and the time series is not over yet and the price continues to move or stagnates. As soon as the timeframe is over, the price suddenly reverses. However, you can hardly prove anything on this forum. At best, you will see a clash of stereotypes and retelling of your ideas. )))
 
The programmer is in shock. Alos will give you a lift...
Yusuf 2=3 ready? Checked it out, is it working?
 
yosuf:

On the other hand, where have we learned to predict the future - nowhere!

If a woman gets pregnant, we can predict that she will give birth in 9 months)))
 
artikul:
Well, if a woman gets pregnant, then we can predict that she will give birth in 9 months)))
Because it is Her Majesty Nature works here, the time belongs to her. If we interfere with our predictions, we will do something wrong.