Correlation indicator. - page 5

[Sergey Vradiy]

Ibragim Dzhanaev:

Wrong.

At least, this style of trading has been described in the literature and has been used for a long time and successfully. However, I would be happy to hear your approach as well.

[Sergey Vradiy]

Ibragim Dzhanaev:

If you try to make a robot, it will fail. If you put indicators in it, it will fail. If you trade a lot of pairs at once, it won't get any better. Proven.

Who says that only correlation should be used in trading? This is not an abolition of the TS, and the correlation may be an element of risk management.

[[Deleted]]

Sergey Vradiy:
At least this style of trading is used by many successful traders. However, I would be happy to know your approach as well.

Type "Bulanov Hedging" into the search engine. It shows not the easiest or safest, but the most profitable way to trade.

The point of trading is to go in on the maximum correlation, when it breaks, that's the profit. I don't trade by levels, like in the video, but trade both ways at once. you get a hedge and a lock. I don't need any indicators. If a global trend is moving up in both pairs, then formally the correlation is there, but in reality the orders will hang, and then by manipulating the lots it is possible to compensate for this error. If the order has gone in the red, for whatever reason, and you do not know how to cut them down, simply close this loss and you will still be in profit for the year. If you close your orders, they will anyway close on the news, even if not the way you wanted initially, even if you are in a small minus. The main thing here is not to be greedy. As for the profit, you cannot earn much anyway. As for the mark, there's no need to be greedy. The market may enter when there is a jamb in correlation, through manipulation with lots (I already wrote).

The correlation is broken, the lock is opened, you earn. And so on.

All the spreads between pairs is nonsense. It has nothing to do with reality. In short, nothing complicated, open the demo and press the button.

Some people might have questions about the loc - it's for safety. Profits are smaller, but if you don't want to, don't lock.

[Timur1988]

Sergey Vradiy:
At least this style of trading has been described in the literature and has been used for a long time and successfully. However, I would be happy to know your approach as well.

Could you give me a link to the literature?

[Timur1988]

Ibragim Dzhanaev:

Type "Bulanov Hedging" into the search engine. It shows not the easiest or safest, but it is more profitable than my way of trading.

The point of trading is to go in on the maximum correlation, when it breaks, that's the profit. I don't trade by levels, like in the video, but trade both ways at once. you get a hedge and a lock. I don't need any indicators. If a global trend is moving up in both pairs, then formally the correlation is there, but in reality the orders will hang, and then it is possible to compensate for this error by manipulating with lots. If the order has gone in the red, for whatever reason, and you do not know how to cut them down, simply close this loss and you will still be in profit for the year. If you close your orders, they will anyway close on the news, even if not the way you wanted initially, even if you are in a small minus. The main thing here is not to be greedy. As for the profit, you cannot earn much anyway. As for the mark, there's no need to do it manually and only when you see the correlation. The market may enter when there is a jamb in correlation, through manipulation with lots (I already wrote).

The correlation is broken, the lock is opened, you earn. And so on.

All the spreads between pairs is nonsense. It has nothing to do with reality. In short, nothing complicated, open the demo and press the button.

Some people might have questions about the loc - it's for safety. Profits are smaller, but if you don't want to, don't lock.

Have you tested the strategy somehow or are you manually trading on it?

[Evgeniy Zhdan]

Sergey Vradiy:
At least, this style of trading has been described in the literature and has been used for a long time and successfully. However, I would be happy to hear your approach as well.

What a load of crap. No approach has been used in forex for a long time and successfully. Especially not the one described in the literature, whatever that may be.

[[Deleted]]

Timur1988:

Have you tested the strategy somehow or are you trading it manually?

By hand.

[Vladimir M.]

Timur1988:

friends, need your opinion. Wrote a Pearson correlation indicator as a line between two pairs and as a table of currency pairs, stocks and indices.

2) As a table in the indicator sub-window in the main chart window

I would like to hear your opinions on the improvement, on the accuracy, presentation and design of these indicators. What may be added or removed? I would be very grateful!

Timur, please make a screenshot of Table 2 for today with this data:

m=48 - number of bars to calculate the correlation between pairs;
period=D1 - daily chart;
price=close - close prices;
bar=0 - 48 bars at close prices starting from zero.

I will compare it with my data. I will post the result.

[Timur1988]

rosomah:

Timur, please make a screenshot, table 2., for today, with this data:

m=48- number of bars to calculate correlation between pairs;
period=D1 - daily chart;
price=close - close prices;
bar=0 - 48 bars at close prices starting from zero.

I will compare it with my data. I will post the result.

please. results for January 12, time 22:05 (Moscow)

[Vladimir M.]

Timur1988:

please. results for 12 January, time 22:05 (Moscow)

I threw a script to check, Just made Alerts (23h.07min) , checked only EURUSD. But we were in a hurry. The results should be identical at the weekend when the close is finally formed. At the same time I checked Spearman and Kendall.

I am pasting the script. Frankly speaking such work is of little use.

//+------------------------------------------------------------------+
//|                                      Correlation_Pearson_etc.mq5 |
//|                        Copyright 2018, MetaQuotes Software Corp. |
//|                                             https://www.mql5.com |
//+------------------------------------------------------------------+
#property copyright "Copyright 2018, MetaQuotes Software Corp."
#property link      "https://www.mql5.com"
#property version   "1.00"

#include <Math\Stat\Math.mqh>

int n_bars=48;
ENUM_TIMEFRAMES enum_t=PERIOD_D1;
//+------------------------------------------------------------------+
//| Script program start function                                    |
//+------------------------------------------------------------------+
void OnStart()
  {
//---
   double arr_1[];
   double arr_2[];
   if(ArrayResize(arr_1,n_bars,0)!=n_bars
      ||ArrayResize(arr_2,n_bars,0)!=n_bars)return;
         
   if(CopyClose("EURUSD.m",PERIOD_D1,0,n_bars,arr_1)!=n_bars
      || CopyClose(Symbol(),PERIOD_D1,0,n_bars,arr_2)!=n_bars)
       return;

   double Pear=0;
   double Spear=0;
   double Kend=0;
   if(!MathCorrelationPearson(arr_1,arr_2,Pear)
      ||!MathCorrelationSpearman(arr_1,arr_2,Spear)
      ||!MathCorrelationKendall(arr_1,arr_2,Kend))return;

   Alert("$ EURUSD.m  &  ",Symbol(),
         "   >>>   Pear=",NormalizeDouble(Pear,4),
         "   >>>   Spear=",NormalizeDouble(Spear,4),
         "   >>>   Kend=",NormalizeDouble(Kend,4));
  }
//+------------------------------------------------------------------+

P.S. The check was done on - Robo, cents. If you want to play with comparisons substitute your symbol for EURUSD.m in the script.