[WARNING CLOSED] UmnickTrader Adaptive EA - 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
What is it adaptable to? Sorry I didn't ask sooner
Adaptation is adaptation, i.e. an EA adapts to changes in the market.
The OOS 9 year test confirms that the EA is capable of adapting successfully over a sufficiently long period of time.
You are a demagogue Artiukhov, in a neglected stage, you should be kicked out of the forum, for flooding and provoking.
https://www.mql5.com/ru/forum/115644
https://www.mql5.com/ru/forum/114665
http://forexsb.com/wiki/start
Gentlemen moderators and philistines, who can tell me what is useful in this thread?
If there is nothing sensible and it certainly was and is flooding, the topic has exhausted itself on the second page ,
Please control more strictly all this nonsense in the thread, which has not left the top of the forum for two months. I am sick and tired of looking at the ramblings.We have a big request to the moderators:
Issue a warning to both sides of the flood (Leo and Artyuhov) and move this thread to read mode (as "newbie-1 questions")
Or delete it as useless for the topic of the forum.
What have you said in this thread that is useful and on-topic?
The code in bold is a modification of its own function.
That is, as you can see, the changes in the code are microscopic.
int countStop = 0;
if( resultTransaction < 0 ) {
// the last trade was losing
arrayProfit[currentIndex] = StopBase;
arrayLoss[currentIndex] = drawDown;
if( countStop > 1 ) {
// change direction of trades
currentBuySell = -currentBuySell;
countStop = 0;
}
else
countStop = countStop+1;
// increase the lot size
absAmount = absAmount+amountStep;
}
The OOS 9 year test confirms that the advisor is able to adapt successfully over a sufficiently long period of time.
So it will continue to be profitable "over a long enough period of time"?
The future is impossible to predict.
However, see the tests in the comments here:
1. 12.05.2010 03:56
Test of a slightly improved version of the "Adaptive Expert Advisor UmnickTrader V3" on the 4.5 year period.
There is only one optimized parameter - StopBase.
Symbol EURUSD (Euro vs US Dollar)
Period 1 Minute (M1) 2006.01.02 00:51 - 2010.05.11 23:59 (2006.01.01 - 2010.05.12)
Model All ticks (the most accurate method based on the smallest available timeframes)
Parameters StopBase=0.005; marketOrderOn=false; spred=0.0005; slippage=200; absAmount=1; timeframe=240;
Bars in history 1453206 Modelled ticks 23538591 Modelling quality 25.00%
Chart mismatch errors 0
Initial deposit 10000.00
Net profit 48460.60 Total profit 277029.60 Total loss -228569.00
Profitability 1.21 Expected payoff 47.32
Absolute drawdown 1902.80 Maximum drawdown 7386.80 (21.01%) Relative drawdown 33.06% (3998.40)
Total trades 1024 Short positions (% win) 508 (53.94%) Long positions (% win) 516 (54.65%)
Profitable trades (% of all) 556 (54.30%) Loss trades (% of all) 468 (45.70%)
Largest profitable trade 524.00 (loss trade -503.60)
Average profitable trade 498.25 losing trade -488.40
Maximal number of continuous wins (profit) 10 (4905.60) continuous losses (loss) 6 (-2963.20)
Maximum number of continuous wins (number of wins) 4905.60 (10) continuous losses (number of losses) -2963.20 (6)
Average continuous gain 2 continuous loss 2
2. Then 31.01.2011 10:38
There were no optimizations in 2010, the parameters used are the same - see the tests below "Adaptive Expert Advisor UmnickTrader V3".
EURUSD symbol (Euro vs US Dollar)
Period 1 Minute (M1) 2010.01.04 00:00 - 2011.01.31 08:31 (2010.01.01 - 2011.02.01)
Model All ticks (the most accurate method based on the smallest available timeframes)
Parameters StopBase=0.005; marketOrderOn=false; spred=0.0005; slippage=200; absAmount=1; timeframe=240; currentIdOrder="1";
Bars in history 382884 Modelled ticks 12330887 Modelling quality 25.00%
Chart mismatch errors 0
Initial deposit 10000.00
Net profit 17674.70 Total profit 82520.10 Total loss -64845.40
Profitability 1.27 Expected payoff 59.51
Absolute drawdown 1214.00 Maximum drawdown 4815.00 (22.07%) Relative drawdown 23.14% (4124.80)
Total trades 297 Short positions (% win) 156 (57.69%) Long positions (% win) 141 (53.19%)
Profitable trades (% of all) 165 (55.56%) Loss trades (% of all) 132 (44.44%)
The most profitable trade is 512.40 (loss trade -507.10)
Average profitable trade 500.12 losing trade -491.25
Maximal number of continuous wins (profit) 9 (4512.10) continuous losses (loss) 8 (-3941.20)
Maximum number of continuous wins (number of wins) 4512.10 (9) continuous losses (number of losses) -3941.20 (8)
Average continuous gain 2 continuous loss 2
As you can see, the tests confirm that the same code was also successful in 2010 on completely unknown new data in advance.
In theory, and preferably then in practice, how do EA results with a positive OOS test at 9 years and a modelling quality of 25% in the tester compare to real world results?
Adaptation is adaptation, i.e. the advisor adapts to changes in the market.
The OOS 9 year test confirms that the advisor is able to adapt successfully over a sufficiently long period of time.
I agree.
What kind of changes?
I agree.
Which one?
You'll laugh, but the councillor "doesn't know" which ones :)
It's about like with this cat - the cat doesn't know what has happened to it either, but is successfully adapting to its new state:
"The experiments were set up this way. In a cat, a part of the extensor (quadriceps femori) on the hind limb was transplanted to the flexor position and, consequently, with this muscle transplantation we obtained a peculiar relation between the centre and the periphery (Fig. 2). Nerve impulses coming from the extensor centre went to both halves of the extensor, because the normal innervation of the two halves of the muscle did not change their relationship to both halves of the muscle. Consequently, the same sending of impulses from the quadriceps muscle centres in one part should have caused flexion and in the other part should have caused extensor.
This circumstance naturally disorganized the entire locomotion of the cat. She made a series of disorganized efforts, extending both hind limbs, then flexing them (Fig. 3). In short, the transplanted part of the muscle was introducing dissonance in the coordination between the flexors and extensors of the limb, but after 1-2 months such disorganisation in the hind limb was over, and the cat walked perfectly normally as if she had not had any muscle transplant.
"