Machine learning in trading: theory, models, practice and algo-trading - page 2901

New comment
 
mytarmailS #:

And in the forex overnight, an order went off,

and look how powerful the level was.


What makes you think it's not a random set of dots?

 
Dmytryi Nazarchuk #:

What makes you think it's not a random set of dots?

It's

 
Aleksey Nikolayev #:

Exactly. If we find an objective definition, it will be a philosopher's stone turning any TC into a grail.

We can ask a more meaningful question, how your guru defines trend and flat.

From what can be seen and formalised. We got extremum from below and above, we set zones around these prices, we wait for extremum from below, if it is in the zone, it can be the beginning of flat, if it is in the zone from above, we can consider that it is flat, the question of waiting time parameter and the size of zones. Besides, there is already a history, where you can practice)))))) But the problem is that more than 3 moves back and forth are rare, and the time of being in flat is random for the most part.

 
Aleksey Nikolayev #:

Wrote a quick seq_list() search for parrtens from sequences, you might be interested in applying it to your data


example data

list_dat <- list()
> for(i in 1:10) list_dat[[i]] <- letters[sample(1:length(letters),sample(3:20,size = 1),replace = T)]
> list_dat
[[1]]
 [1] "w" "g" "m" "t" "n" "p" "g" "d" "m" "l" "y"

[[2]]
 [1] "t" "j" "i" "l" "h" "g" "y" "o" "s" "q" "k" "g" "s" "w" "z"

......

.....

....

pattern

pattern <- c("f","b","k")


search

seq_list(pat = pattern, list_dat = list_dat)

 [1] FALSE  TRUE FALSE FALSE FALSE FALSE FALSE FALSE FALSE  TRUE


list_dat
[[1]]
 [1] "e" "j" "y" "e" "t" "m" "a" "y" "w" "y" "s" "l" "n"

[[2]]
 [1] "q" "x" "r" "u" "a" "f" "d" "y" "q" "b" "n" "k" "n" "b" "j" "s" "t" "s" "n" "l"

[[3]]
[1] "b" "f" "x" "l" "u" "h"

[[4]]
 [1] "z" "m" "g" "p" "g" "h" "s" "y" "u" "y" "u" "r" "z" "a" "b" "r" "b"

[[5]]
 [1] "v" "u" "l" "r" "h" "c" "e" "o" "c" "a" "p" "j"

[[6]]
[1] "r" "a" "b" "r"

[[7]]
[1] "o" "a" "k" "c" "d" "e"

[[8]]
[1] "s" "j" "g" "l"

[[9]]
[1] "q" "b" "x" "v" "f" "s" "x"

[[10]]
 [1] "v" "n" "f" "i" "a" "l" "b" "a" "w" "k" "a" "g"


==========================

function code

char_seq_list <- function(){
  library(Rcpp)
  src <-
    "LogicalVector char_yes_seq_vec_cpp(CharacterVector pat, List list_dat){
  
  LogicalVector res_vec( list_dat.length() , false );
  
  
  
  for(int i=0; i<list_dat.length(); ++i){
  
  LogicalVector pat_vec (pat.length() ,false); 
  
  
  
  int cnt_end_pat = 0;
  CharacterVector List_vec = list_dat[i];
  
  
  for(int j=0; j < List_vec.length(); ++j){
  
  if(List_vec[j]==pat[cnt_end_pat]){
  pat_vec[cnt_end_pat] = true;
  cnt_end_pat = cnt_end_pat+1; 
  }
  if(cnt_end_pat==pat.length())  break;
  }
  
  res_vec[i] = is_true(all(pat_vec==TRUE));
  
  }
  return res_vec; 
}"

char_yes_seq_vec_cpp_list <<- Rcpp::cppFunction(src)
}

seq_list <- char_seq_list()
 
mytarmailS #:

The discount rate changes every n-minutes??

It's hard to say - I haven't identified exactly how it changes - most often due to stronger movement, according to my observation. The point is the constant skewing of the day's results.

 
mytarmailS #:

I don't know, I'm not gonna argue.

Here again you can see that the price on the CME after the entry went where it should not have gone....

in the red squares "miss-clicks" on the glass, it shouldn't have.


I never thought it would be more pleasant to trade at Forex.

By the way, is this the nearest expiry futures traded?

Where is the spread bigger?

 
Aleksey Vyazmikin #:

Hard to say - I haven't identified exactly how it changes - most often due to stronger movement, in my observation. The point is the constant skewing by the result of the day.

The bet is designed to have a long-term or medium-term impact on a particular country's market. On daily charts, the bet is not obliged to fulfil the wishes and desires of speculators.

The currency market moves with the economy. The speculative factor is of mediocre importance. Liquidity in the form of speculators is the main factor in the currency market on daily charts.

MO should be built on new factors. it is necessary to solve new problems, and not to follow the outdated footsteps of sensais)))

 
Uladzimir Izerski #:

MoD should be built on new factors. it is necessary to solve new problems, not to follow the outdated footsteps of sensai))))

To begin with, you need to have some idea of the market, to have some model, otherwise there are endless variants, it's like fighting with the shadow...

Well, following the mainstream, copying what was written in blogs pimply mum's python datasaintists is stupid....
That's for Maximka.)

It's not thinking, just following others, following the herd, being in the herd ...
 
mytarmailS #:
To begin with, you need to have some idea of the market, to have some model, otherwise there are endless options, it's like fighting with the shadow....

And following the mainstream, copying what pimply mum's python datasaintists wrote in their blogs is stupid....
That's for Maximka.)

Here we come to the starting point.

Where from, where to, why? I didn't formulate it.

""any idea of the market?"" - It's the same story on the charts :) The smirk is harmless.

 
Uladzimir Izerski #:

The rate is designed for long-term or medium-term influence on the market of a particular country. On daily charts, the rate is not obliged to fulfil the wishes and desires of speculators.

What was the point of those words? I was talking about the fact that the rate is embedded in the futures contract and at least once a day it is adjusted relative to the Forex/spot according to the results of trading.

Uladzimir Izerski #:

The foreign exchange market moves with the economy. The speculative factor is of mediocre importance. Liquidity in the form of speculators is the main factor in the currency market on daily charts.

Everyone considers the value of a currency when making buy or sell decisions - whether it is a tourist or a large trading company.

MO should be built on new factors. we need to solve new problems, not follow the outdated footsteps of sensai)))

What exactly, "new" factors?

What "new" challenges do you propose to solve?

Without specifics it looks more like party slogans.

1...289428952896289728982899290029012902290329042905290629072908...3574
New comment