Discussion of article "Probability theory and mathematical statistics with examples (part I): Fundamentals and elementary theory" - page 6
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
Not from time. From changes in the properties of the market we want to exploit.
The speed of response should, to a first approximation, match the speed of the events to which the system is responding.
If the reaction speed is higher, good. If it is lower - bad, we do not have time to react.
I.e."sometimes" is different and depends on a particular system.
"Butter is butter." Market properties change over time. Some algorithm of yours changes the parameters of the original system according to those properties and they become functions of time through dependence on those properties. If we have functions y=y(x) and x=x(t), then the function y=y(x(t)) will already be a function of time. Your algorithm for changing the parameters of the original system obviously needs tuning (reaction rate, for example), which is usually done by introducing some parameters for it and selecting them. These parameters will be the parameters of the new system built from the original one.
let's talk terminology.
If we are talking in the context of price prediction, then we must necessarily take into account the reliability of the prediction
how will the forecast error be evaluated?
3. is it item 1 + item 2 ?It depends more on the specific TS, somewhere time is more important, somewhere price, somewhere both factors.
The equity of the TS is the evaluation of the price forecast. And it is better to make it (evaluation) not only by closing a deal, but also by tacts inside an open deal, which is what I am talking about above).
Igor Makanu:
it doesn't work that way.
let's ignore capital management, although it is a significant part of the TS.
for trading it is important not only to predict where the price will reach, but it is also important to assume when it will happen, or it is necessary to take into account what happened before the moment of market entry, i.e. it is necessary to analyse a certain set-up or pattern.
but just to make a forecast that the price will break such and such level... Well, everyone does it on thematic forums, but rarely these forecasts help to earn money.
I agree if you assume a forecast for one bar and every bar to close and open a deal, but such TS will work on TF D1 and above.
Well, yes, "what happened before the moment of market entry" is a price analysis, on the basis of which the forecast of further behaviour after the set-up is made (whether to open a trade or not).
About "one bar" is to simplify the discussion, well, you can switch from D1 to H1, and "forecast for 1 bar" will be called "forecast for 24 bars"."Oil of Oil. Market properties change with time. Some algorithm of yours changes the parameters of the original system according to these properties and they become functions of time through dependence on these properties. If we have functions y=y(x) and x=x(t), then the function y=y(x(t)) will already be a function of time. Your algorithm for changing the parameters of the original system obviously needs tuning(reaction rate, for example), which is usually done by introducing some parameters for it and selecting them. These parameters will be the parameters of the new system built from the original one.
The main idea is that the astronomical time scale is linear, but market properties usually change extremely non-linearly. And more often than traders recalculate the model.
The idea about the reaction speed is that the speed of the exploited events is usually known from history and is approximately constant (the same gaps, for example), it is not necessary to constantly adjust to it, it is enough to choose the reaction speed once with a reserve.
I think all this is clear to you without me) I just feel that textbooks on SLUPs, and even more so on matstat, avoid these nuances.
The basic idea is that the astronomical time scale is linear, and market properties usually change in a highly non-linear fashion. And more often than traders recalculate the model.
The idea about reaction speed is that the speed of exploited events is usually known from history and is approximately constant (the same gaps, for example), it is not necessary to constantly adjust to it, it is enough to choose the reaction speed once with a reserve.
I think all this is clear to you without me) I just feel that in textbooks on SLUPs, and even more so on matstat, these nuances are bypassed.
Time in SLUPs is an extremely abstract object. Perhaps, we can only say that it is one-dimensional and it is impossible to look into the future.
I am only disagreeing with your initial statement about the inapplicability of piecewise constant models and claiming that (from a very general point of view) we only use them. Roughly speaking, no matter what ideas we come from - we will end up with an algorithm with a set of parameters (which are constant or sometimes change). There is only one alternative to this - haphazard trading. Of course, there is no practical use of this statement, but that does not make it wrong.
It depends more on the particular TC, somewhere time is more important, somewhere price, somewhere both factors.
Equity of the TS is an estimation of price forecast. And it is better to make it (evaluation) not only by closing a deal, but also by tacts inside an open deal, which is what I am talking about above).
;)
norms
@Aleksey Nikolayev explain "on your fingers"...preferably without showing the middle one ))
example 4 from the article
if after calculation by script I got the following results:
n1e=31 p1e=0.7741935483870968 p2e=0.499786097834425
what do these values show against your calculated values from the article:
Ответ: n1e = 21; p1e = 0.71; p2e = 0.17 (числа округлены). Он кажется достаточно очевидным − наша модель "увидела" смену направления (или коррекцию) движения цены в конце её участка. Это говорит о том, что переход к более сложной модели в данном случае был не напрасен.
The basic idea is that the astronomical time scale is linear, and market properties usually change in a highly non-linear fashion. And more often than traders recalculate the model.
The idea about reaction speed is that the speed of exploited events is usually known from history and is approximately constant (the same gaps, for example), it is not necessary to constantly adjust to it, it is enough to choose the reaction speed once with a reserve.
I think all this is clear to you without me) I just feel that in textbooks on SLUPs, and even more so on matstat, these nuances are bypassed.
On the one hand, you can systematically (and quite reasonably) move to non-linear and curved scales of price and time... (with time it's a mess, the density jumps back and forth).
and otherwise, in the classics of trading there are only two engineers - you can see the skill, technicians don't lie.
the fundamentals! Everything else is superficial and added by managers.
@Aleksey Nikolayev explain "on your fingers"...preferably without showing the middle one ))
example 4 from the article
if after calculation by script I got such results:
n1e=31 p1e=0.7741935483870968 p2e=0.499786097834425
What do these values show against your calculated values from the article:
Try running the script "discret_prices.mq5" on the same chart and with the same parameters (as you ran the script "p1p2_model.mq5") and see if the picture of the discretised price is similar to mine (and the sequence of 1's and -1's that is given in the article before the EURUSD chart). I got n=27 at all, and it should be greater or equal than n1e
Try to run the script "discret_prices.mq5" on the same chart and with the same parameters (as you ran the script "p1p2_model.mq5") and see if the picture of discretised price is similar to mine (and the sequence of 1 and -1, which is given in the article before the EURUSD chart). I got n=27 at all, and it should be greater or equal than n1e
the chart is different, I'm just dealing with the provided mathematical apparatus.
that's why the question arose, what do we get if according to Bernouli's scheme my values
for this graph:
n1e=34069 p1e=0.5006604244327688 p2e=0.09090909090909091
D1 chart itself, where the trend is clearly present