Discussion of article "Extract profit down to the last pip" - page 12

New comment
 
aleger:

It's not just the "most minimal price movements" that are interesting, but the total (including minimal) durations of those movements...

You arewelcome (values on the right are weekend moves). There are quite a lot of analysis services.

 
fxsaber:

Please (the values on the right are weekend passages). There are quite a lot of analysis services.

This is data for which "instrument"?

 
aleger:

This is data for what "instrument"?

Updated data from the account frozen in the article.

 
fxsaber:

Updated data from the account monitored in the article.

Thank you. I wish I could really learn/adapt to scratch out a profit in full - from the beginning to the end of every profitable current trend/movement!

 
fxsaber:

Please (the values on the right are weekend passages). There are quite a lot of analysis services.

Interesting monitoring. You can take it as a sample for presenting trading results.

It is a pity that it is only Forex. And there is no interactive analysis (as far as I understand).

 
Dmitriy Skub:

There is nointeractive analysis (as far as I understand it).

What is it?

 
fxsaber:

What's that?

For example, investigate the parameters of the TS depending on on/off trades on all Mondays/Tuesdays, etc. Or hours within a day.

Etc.

 
Dmitriy Skub:

For example, to investigate the parameters of the TS depending on on/off trades on all Mondays/Tuesdays, etc. Or hours within a day.

Etc.

It's all possible.


For example, profit distributions for each of the TCs that have been run. They haven't changed since the launch.

 

This is how you can see that it was undesirable to trade on Thursdays. But such analyses are, of course, better done on backtests.


And to find the best intervals of the day on your account, including the stock exchange, you can use such a script.

Forum on trading, automated trading systems and testing trading strategies.

Libraries: BestInterval

fxsaber, 2018.10.12 16:38

#include <fxsaber\BestInterval\BestInterval.mqh> // Calculation of the best trading interval

void OnStart()
{
  BESTINTERVAL BestInterval; // Created an object to calculate the best trading interval
  
  // Array of closing deals
  DEAL Deals[];
  
  // We need to write two fields for each closing transaction
  // Deals[i].Profit - profit of the deal closing the position
  // Deals[i].OpenTime - time of opening (not closing) the position that the deal closes
      
// BestInterval.Set(); // Put trading history - uses MT4Orders
  BestInterval.Set(Deals); // Posted a self-prepared bidding story
  
  const int AmountIntervals = 3; // How many worst-case trade intervals to throw away
  
  for (int i = 0; i < AmountIntervals; i++)
    if (BestInterval.DeleteWorseInterval()) // If something was thrown away
      Print(BestInterval.ToString());       // Let's print out the obtained trade data
    else
      break;                                // Otherwise, we're out
}

That is, you can whatever data you want to feed in to calculate the best interval. The library is platform independent.


The result of running this script

SUMMARY: 00:00:00 - 23:59:59 : Profit = 672.54 (100.00%), Total = 1680 (78.87%), PF = 2.43, Mean = 0.40, DD = 60.68, RF = 11.08
22:59:47 - 23:59:59 : Profit = 151.63 (22.55%), Total = 222 (86.49%), PF = 3.47, Mean = 0.68, DD = 32.61, RF = 4.65
21:59:57 - 22:47:36 : Profit = 78.94 (11.74%), Total = 269 (78.81%), PF = 2.15, Mean = 0.29, DD = 33.47, RF = 2.36
21:11:48 - 21:59:37 : Profit = 109.40 (16.27%), Total = 309 (85.11%), PF = 2.55, Mean = 0.35, DD = 26.98, RF = 4.05
00:00:00 - 21:01:31 : Profit = 332.57 (49.45%), Total = 880 (74.77%), PF = 2.23, Mean = 0.38, DD = 29.87, RF = 11.13
Amount of Delete Intervals = 3 (2019.07.25 - 2019.09.25)

SUMMARY: 00:00:00 - 23:59:59 : Profit = 642.76 (100.00%), Total = 1692 (78.55%), PF = 2.28, Mean = 0.38, DD = 82.89, RF = 7.75
22:59:47 - 23:59:59 : Profit = 151.63 (23.59%), Total = 222 (86.49%), PF = 3.47, Mean = 0.68, DD = 32.61, RF = 4.65
21:11:48 - 22:47:36 : Profit = 158.56 (24.67%), Total = 590 (81.19%), PF = 1.93, Mean = 0.27, DD = 58.69, RF = 2.70
00:00:00 - 21:01:31 : Profit = 332.57 (51.74%), Total = 880 (74.77%), PF = 2.23, Mean = 0.38, DD = 29.87, RF = 11.13
Amount of Delete Intervals = 2 (2019.07.25 - 2019.09.25)

SUMMARY: 00:00:00 - 23:59:59 : Profit = 584.95 (100.00%), Total = 1842 (76.60%), PF = 1.95, Mean = 0.32, DD = 117.17, RF = 4.99
22:59:47 - 23:59:59 : Profit = 151.63 (25.92%), Total = 222 (86.49%), PF = 3.47, Mean = 0.68, DD = 32.61, RF = 4.65
00:00:00 - 22:47:36 : Profit = 433.32 (74.08%), Total = 1620 (75.25%), PF = 1.78, Mean = 0.27, DD = 105.41, RF = 4.11
Amount of Delete Intervals = 1 (2019.07.25 - 2019.09.25), 23:00 - 23:00, CountHours = -1

SUMMARY: 00:00:00 - 23:59:59 : Profit = 455.60 (100.00%), Total = 1897 (75.70%), PF = 1.59, Mean = 0.24, DD = 257.31, RF = 1.77
00:00:00 - 23:59:59 : Profit = 455.60 (100.00%), Total = 1897 (75.70%), PF = 1.59, Mean = 0.24, DD = 257.31, RF = 1.77
Amount of Delete Intervals = 0 (2019.07.25 - 2019.09.25)

It shows that you should not have traded from 22:47 to 23:00. But this is most likely a fitting, no forward analysis was done. Especially since it is correct to get rid of MM influence first - just count everything in your favourite pips.

 

Then there are no questions - they have provided everything that is necessary.

I agree about counting in pips. I wonder how to get it into the heads of MT developers, who have been counting MO in the deposit currency for 30 years.

However, I don't use the tester - maybe something has changed?

1...5678910111213141516171819...36
New comment