Experiment
Dear forum users. I decided to discuss the results of my experiment on the organization of real automatic trading with one order without SL and TP on EUR/USD currency pair on TF M1 on cent account with a minimum lot 0.01, by analyzing 1000 last minute bars of history that I started today - 23.04.21:
| Por no. | opening | status | price | SL | TP | closing | price | P/S | Total U/U |
| 1 | 03.27.29 | sell | 1.2022 | 0 | 0 | 03.35.23 | 1.2025 | -0.2 | -0.2 |
| 2 | 0.3.35.53 | buy | 1.2025 | 0 | 0 | 03.43.02 | 1.2023 | -0.2 | -0.4 |
| 6 | 03.49.10 | sell | 1.2022 | 0 | 0 | 03.53.01 | 1.2024 | -0.2 | -0.6 |
| 4 | 0.3.53.02 | buy | 1.2024 | 0 | 0 | 04.04.29 | 1.2023 | -0.1 | -0.7 |
| 5 | 04.04.30 | sell | 1.2022 | 0 | 0 | 03.24.02 | 1.2021 | 0.1 | -0.6 |
| 6 | 0.4.25.03 | buy | 1.2022 | 0 | 0 | 16.53.10 | 1.2049 | 2.7 | 2.1 |
| 7 | 16.53.11 | sell | 1.2049 | 0 | 0 | 17.24.18 | 1.2061 | -1.2 | 0.9 |
| 8 | 17.24.19 | buy | 1.2061 | 0 | 0 | 1.2095 | 3.4 | 4.3 |
Preliminary conclusion: With minimal losses, within the spread, it is possible, despite the price fluctuations within certain limits, to cut off the flat and attack the trend track. The experiment is continuing. In the future I want to arrange similar trading on all 34 instruments available to me, including gold and silveron VPS. Your opinions.
There is very little data to draw any conclusions. There are three states in the market, if the robot is untested for all three states, it is likely to go astray in any of them.
I agree, there is not enough data yet, but the experiment will show, if the stated way of trade organization is correct. I will try different sampling up to 1000 history bars and up to 10000 bars. I do not understand why the robot should or may go astray. It clearly controls the price trend and changes the order status in time. There is no reason or chance for it to go astray.
Greetings!)
Why 1,000?
How about 2,000? Or 5,000?
What is the algorithm based on?
- No, I haven't.
Greetings!)
Why 1,000?
How about 2,000? Or 5,000?
What is the basis of the algorithm?
Indeed, so far, it's a randomly taken figure. I think it should, roughly, have that order. There can be 2000, 30000, ...., 10000, for each pair we need to find, ideally, as Dmitry suggests https://www.mql5.com/ru/forum/367874#comment_22007136, on the tester. The algorithm is based on the predictive capabilities of my indicator.
Totally agree. It's the best way to find the optimal sample. It takes time. I'll look into it at the same time.
sorry, can you show me the indicator?
On the screenshots on the mt4 chart.
Thank you.
Indeed, so far, it's a guessing figure. I think it should, roughly, be in that order. The variants 2000, 30000, ...., 10000 are possible, for each pair should be found, ideally, as Dmitry suggests https://www.mql5.com/ru/forum/367874#comment_22007136, on the tester. The algorithm is based on the predictive capability of my indicator.
It makes no difference, as soon as the number of samples starts to depend on the user, immediately the strategy test automatically turns into a fitting to the story.
It makes no difference, as soon as the number of counts starts depending on the user, immediately the strategy test automatically turns into a story fit.
I understand the phrase with a very deep understanding. It is commendable.
But I can't understand it to the end. What is the beginning... Where is the beginning?
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
You agree to website policy and terms of use
Dear forum users. I decided to discuss the results of my experiment on the organization of real automatic trading with one order without SL and TP on EUR/USD currency pair on TF M1 on cent account with a minimum lot 0.01, by analyzing 1000 last minute bars of history that I started today - 23.04.21:
Por no.
opening
status
price
SL
TP
closing
price
P/S
Total U/U
1
03.27.29
sell
1.2022
0
0
03.35.23
1.2025
-0.2
-0.2
2
0.3.35.53
buy
1.2025
0
0
03.43.02
1.2023
-0.2
-0.4
6
03.49.10
sell
1.2022
0
0
03.53.01
1.2024
-0.2
-0.6
4
0.3.53.02
buy
1.2024
0
0
04.04.29
1.2023
-0.1
-0.7
5
04.04.30
sell
1.2022
0
0
03.24.02
1.2021
0.1
-0.6
6
0.4.25.03
buy
1.2022
0
0
16.53.10
1.2049
2.7
2.1
7
16.53.11
sell
1.2049
0
0
17.24.18
1.2061
-1.2
0.9
8
17.24.19
buy
1.2061
0
0
1.2095
3.4
4.3
Profit and loss as a function of time in hours since the start of the ATC
The preliminary conclusion: With minimal losses, within the spread, despite the price fluctuations within certain limits, it is possible to cut the flat and attack the trend track. The experiment is continuing. In the future I want to arrange similar trading on all 34 instruments available to me, including gold and silveron VPS. Your opinions.