Shall we discuss an expert based on coding? - page 3

New comment
 
Gutman:

I don't see anything trend following in this system, just a stupid indicator pattern with positive statistics, no guarantees, no trend, we have no idea what the trend is at each current moment of code arrival

The absence of the notion of a trend in your system is a big omission, if only for the reason that big profits are only in the trend and everything else is irrelevant. I am sure that if you add the definition of a general trend to your system, for example from large charts of 4 - 8 hours, and open only in the direction of the trend, the output results will improve considerably.
 
VNIK:
The lack of trend definition in your system is a big omission, if only for the reason that big profits are only in trend and everything else is irrelevant. I am sure that if you add the definition of a general trend to your system, for example from large charts of 4 - 8 hours, and open only in the direction of the trend, then the output results will improve significantly.
Of course there is such an element as a filter (request to the higher frame and writing it into the code), and after adding it the results have naturally improved, but still I do not consider it as following the trend
 

" adjusting to history".

We are trying to predict the future based on the past, everybody does that, and as for describing the market, the indicator performs this task, it smoothes it out, tries to identify trends and patterns, after all, all indicator experts work with an event on the indicator, we have tried to describe the maximum number of events and used them for statistics, but how else?

 
papaklass:
Can you lay out your classification of markets? How many types of markets are there in your model? Otherwise we are getting into the thick of things.
we classify roughly: bullish or bearish and how many waves it has already been that way, where the current wave is going in this or that market at the moment, and what the wave is doing in the higher timeframe
 
I never realised that mooing was as unpleasant as mooing.
 
sergeev:

I never realised that it turned out to be as unpleasant to mooch as it is to mooch.
When you speak impersonally, it is more a claim to unquestionable truth, we want to be criticized, knowingly admitting our mistakes, that is, we are talking about our personal not claiming to be true, give an example of a statement in impersonal form and compare
 
"we want..."
it's still unpleasant. :)
 
sergeev:
"we want..."
it's still unpleasant. :)
(us - remove) I would like to hear from the State moderator something on the subject (the transition to personality) is not contrary to the rules of the forum?
 
Gutman:
we classify roughly: bullish or bearish and how many waves it already is, where the current wave is going in this or that market at the moment, and what the wave is doing in the higher timeframe

It would not be a bad idea to first define the problem statement and decide for yourself what you want.

First you write that you don't define a trend as such. Two posts below, the wave markup appears. And this is without defining a trend?

 
VNIK:

It would be a good idea to start by defining the problem and deciding for yourself what you want.

First you write that you don't define a trend as such. Two posts below, the wave markup appears. And this is without defining a trend?

The point is that this is not the purpose - to determine the trend, etc., it's just applied to the coding algorithm, then statistics, and the surprising thing is that if you think "trend", the results of these codes are 50/50 gray, but if purely based on statistics, there are codes that are not related to trends in any way and generally by eye this would not enter, but the expert and statistics speak forwards and take its place, so how can I say that we work based on trends?
1234567
New comment
Reason: