Once again, it's about the eternal: trend/flat. - page 19
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
1. For God's sake, as they say. If it helps you to improve your system's performance with an understanding of the market situation as "volatility" - go for it!
2. Formalisation - the ability to describe it numerically and formulaically. This I have given.
3. I showed the results of my same flat system over 2 years, 1 year of which is optimization. Everything is correct. Have a look above on this page.
2. would you be kind enough to repeat it.
3. Correct - when the optimized section is not involved at all. Only a test one. (Besides, one cycle is not enough.)
1. be kind enough to repeat it.
2. Correct - when the optimised section is not involved at all. Only a test one. (Also, one cycle is a bit much.)
1. Repeated on page 17, briefly achieved in this thread.
2. Not involved in what? 2014.01.01-2015.01 optimization section, then test 2014.01.01-2016.10.01 so that one can clearly see on one graph the optimization section and the unfamiliar section.
1. Repeated on page 17 briefly the achievements of this thread.
2. Not involved in what? 2014.01.01-2015.01.01 optimization section, then test 2014.01.01-2016.10.01 so that the optimization section and the unfamiliar section are clearly visible on the same graph.
3. You have an optimization plot 2014.01.01-2015.01.01. Then the test plot should be 2015.01.01-2016.10.01 And you should compare results for that year.
Better yet Optimization 2012 - Test 2013, Optimization 2013 - Test 2014, Optimization 2014 - Test 2015, etc. And the sum of the test plots is compared. (if you decide that a yearly interval is better for your system)
3. You have optimization section 2014.01.01-2015.01.01. Then the test one should be 2015.01.01-2016.10.01 And you should compare results for that year.
Better yet Optimization 2012 - Test 2013, Optimization 2013 - Test 2014, Optimization 2014 - Test 2015, etc. And the sum of the test plots is compared. (if you decide that the annual interval is better for your system)
Look at the two screenshots from the tests. Note why exactly two, and how they differ. Think about why these particular screenshots are the same.
I will give you a hint - to compare the system operation with and without a flat filter. And not to compare optimization and test sections. Otherwise I would have done so - two screenshots of optimization and test sections. Fucking hell...
Look at the two screenshots from the tests. Pay attention to why there are two, and how they differ. Think about why these two screenshots are different.
I will give you a hint: I want to compare the system operation with and without a flat filter. And not to compare optimization and test sections. Otherwise I would have done so - two screenshots of optimization and test sections. Fucking hell...
And here is a comparative table of check forwards for time and volatility splits. The step of the wolf forward is 1 month, optimization is 2 months. These proportions have been selected by preliminary testing, i.e. they are optimal for this indicator and this pair. The initial sets are the same for each EA, by the way, the experiment has to be pure. The optimization method is enumerating all parameters, no genetics.
Month
Pprofit
Oct
98,9
-281,4
November
-96,2
256,8
Dec
186,7
712,7
Jan
785,4
401,9
Feb
-255,7
-133,9
March
-182,8
174,4
April
424,9
312,9
May
327,7
459,1
June
-249,8
-215
July
697,7
330,8
Aug
195,5
-119,7
September
91,2
212,9
Total:
Total:
2023.5
2111.5
First Expert Advisor -WPR 2pc + time division by day and night, trades 24 hours Total 4 variables
Second EA WPR 2pc + time division by ATR, trade 24 hours Total 5 variables (1 added for ATR, it describes a shift to the past to define whether volatility decreases or increases, timeframe and period are the same as for one of wpr).
There is no difference.
Unfortunately, I was not able to extract formalizations about trend-flat from your links, and, accordingly, to check them.
Conclusion - the difference between day and night trading on the oscillator, as a rule, is not related to the trend or its absence, but to the general volatility of the market, which is, of course, lower at night.
Andrey Dik, what is your system's window for determining a flat (to me it is not a flat, but a so-called shelf - a flat in my understanding is a more volatile entity), is it automatically determined or is it found by optimisation?
The first Expert Advisor -WPR 2pc + time division by day and night; we trade 24 hours Total 4 variables
Second EA WPR 2pc + split by ATR, trade 24 hours Total 5 variables (1 added for ATR, characterises a shift to the past to determine whether volatility is decreasing or increasing, the TF and period is exactly the same as one of the wpr).
There is no difference.
Unfortunately, I was not able to extract formalizations about trend-flat from your links, and consequently check them.
The conclusion is that the difference between day and night trade by oscillator is not usually related to the trend or its absence, but to the total market volatility which is naturally lower at night.
What makes you think I have an oscillator in my system?
And what do you understand by the word "volatility"? And how do you know if the volatility is suitable for trading now and not?
Window?