Building a trading system using digital low-pass filters - page 28
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
to Neutron
It may be closer to quantum mechanics, but then where is the method of constructing the Schrödinger equation ?
Or, would be happy to apply methods of self-consistency, and, on semiempirical constants agree in advance, but what to agree with what ?
..... let's write down the Eigenfunction of Price Movement)))
The latter, I would say figuratively and imprecisely, are closer in their internal logic to quantum mechanics or something... can only be operated on statistical material with aggravating additions in the form of all sorts of non-stationary phenomena.
and market participants are closer to what are like gravitons or phonons? )
A joke about a statistician
Doctor: was the patient sweating before he died?
Nurse: sweating.....
Doctor: -Good!
___
The Market Model for Digital LF Filters
"Smart fleas on a fat dog"
or like the fleas who know nothing about the dog's life, but have a low-pass filter,
to guess the movements of the host and thereby profit from the host's movements.
-Dog asleep
-Dog eats
-The dog itches (do not enter)
-The dog wags its tail (no fleas to bite)
-Dog breeds (you can change the carrier)
-Dog catches fleas (stages: dangerous pause before snapping teeth, fine-pitch plucking)
=All these tasks are solved by a good low-pass filter.
and market participants are closer to what are like gravitons or phonons? )
To a population of charged particles.
The row is one, but there are many candlesticks in it, and when there are many candlesticks, it's already a statistic.
It's about distribution of candlesticks' heights. To be more precise, it means that very big candlesticks are much bigger than they should be if no one is standing behind them.Hmm... So the problem is formulated (in particular) as a prediction of the next candlestick based on a series of previous values?
About big candlesticks... What was the assumption made that there shouldn't be many of them? I mean the shape of the distribution curve?
There are zeros along the edges anyway, i.e. there are no candlesticks of + or - a million points and more. But the curve can be of any kind, even with one maximum or with two or more?
It is clear that there are a lot of big candlesticks because at certain times (like news) the market performs coordinated one-way movements, and at other times there is no such coordination.
We just need an appropriate model.
I explained to jartmailru what "digital filters" are
and jartmailru is aware of what filters are.
.
I needed clarification regarding my question.
https://forum.mql4.com/ru/10977/page28
(which basically boils down to - whether fx.qrz.ru has errors - or not)
.
>> christmas, you are answering me, but not to my questions - but to whatever you want -
Don't answer me again, please.
and jartmailru is aware of the fact that there are filters.
.
I needed clarification regarding my question
https://forum.mql4.com/ru/10977/page28
(which actually boils down to whether fx.qrz.ru has errors or not)
.
christmas, you are answering me, but not to my questions - but to what you want -
Don't answer me again, please.
your questions are as stupid as they are lengthy in text
the author of http://fx.qrz.ru/ (my countryman by the way) lives in Novosibirsk
for your information:
Novosibirsk is world-famous for the Novosibirsk Academgorodok, which is home to dozens of scientific research institutes
create your own filter generator "without errors" if you do not like this one.
SZS by the way I answered your question long time ago https://forum.mql4.com/ru/10977/page26
But you have not been complacent and continue to demonstrate your knowledge of filters and blame the generator
jartmailru wrote >>
Question itself:
Is there no problem with these digital filters?
no
and jartmailru is aware of the fact that there are filters.
I needed clarification regarding my question
https://forum.mql4.com/ru/10977/page28
(which basically boils down to whether - or not - fx.qrz.ru has errors)
Surely there are no errors there, otherwise they would have been identified long ago.
I have not worked with this program, so I can not say specifically on it.
In general terms - the filter should pass frequencies within its bandwidth, and crush what's beyond it.
So, if you feed low frequency, it should pass through, and if you feed high frequency, it should crush.
And watch without subtraction - just before and after filtering. Should work according to theory.
your questions are as silly as they are lengthy in text
By the way I answered your question a long time ago https://forum.mql4.com/ru/10977/page26
So don't answer :-). Show some dignity.
.
The question on page 26 is a great one - viz.
If Out1 = Filter1(Signal), why
Out2 = Filter1(Signal - Out1) is so big. and why
Out3 = Filter1(Signal - Out1 - Out2) is so big...
.
A really stupid question was asked on page 27.
Namely: "why do LowPass filters 10...12, 14...16, 20...22 .... 44...48 are missing signal period 50".
the answer is obvious: because 50 is in their bandwidth.
.
On page 28 I asked why a 44...48 filter skips a signal with a period of 50.
(i.e. with distortion of both amplitude and phase).
.
What do you do for a living?
.
I don't care if it's a cluster strategy tester. If only for the sake of it.
Read here for answers to all your questions.