Discussion of article "Self-adapting algorithm (Part IV): Additional functionality and tests" - page 3
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
As the number of instruments increases, the equity drawdowns will decrease. I have shown an example of how it works in Figure 9. If you test one instrument from 2008, the drawdowns will be much higher, dozens of times, and this is partly the point of the algorithm. After I upgrade the current version to full-fledged work, I plan to expand the number of traded instruments up to a hundred. And there is a lot to improve, a lot of mechanics can be improved. Whether or not it is an overstayer is not important. It is much more important that having a model you can calculate maximum drawdowns theoretically and forecast further work.
Maxim, the increase of instruments can only "deceive" you, it will show a smoothed Equity on the existing history. But in real life synergy will easily become dissynergy, because you have not investigated it in any way. And Murphy's law will work, negative influences will multiply.
Maxim, the increase of instruments can only "deceive" you, it will show a smoothed Equity on the existing history. But in real life synergy will easily become dissynergy, because you have not investigated it in any way. And Murphy's law will work, negative influences will multiply.
What would be the difference between the real and the test?
Have you overlaid the test results on each other in time? Synchronise Equity/Free Funds/Balance over time, what will you get even with tests simultaneously on the account for the whole portfolio of instruments?
Example from real life, an acquaintance trader on the American stocks very successfully segmented the market and formed a portfolio, successfully traded it at the end of the year, at the opening of the new year, almost all of his portfolio was made up of "leaders of the fall". I don't remember the % losses.
Have you overlaid the test results on each other in time? Synchronise Equity/Free Funds/Balance by time, what will you get even with tests simultaneously on the account for the whole portfolio of instruments?
Example from real life, an acquaintance trader on the American stocks very successfully segmented the market and formed a portfolio, successfully traded it at the end of the year, at the opening of the new year, almost all of his portfolio was made up of "leaders of the fall". I don't remember the % losses.
So it's a test of 28 instruments at once. So yes, all the equity is already overlapped.
We are not talking about correlation, but just a systemic effect. If you are sure that everything is taken into account when working in the "pile", then, - successful reals!
It's not correlation, it's just a systemic effect. If you are sure that everything is taken into account when working in the "pile", then, - have a successful real!
It is too early for this robot to go real, first it needs to raise profitability and reduce drawdown.
Perfectionism in a situation where there is only one in three persons (customer, critic, implementer) is more harmful in my experience. Before the implementation and already obtaining the real result at least negative (does not work) or positive (profit, experience) do not reach quite satisfactory intermediate versions. If there is a technical possibility and time will take minimum - put to test on demo.
Perfectionism in the situation - one in three persons (customer, critic, implementer) is more harmful in my experience. Before implementation and already obtaining the real result at least negative (does not work) or positive (profit, experience) do not reach quite satisfactory intermediate versions. If there is a technical possibility and time will take minimum - put to test on demo.
The robot is written for real accounts and of course I tested it on demo, the trades fully coincide with the tester, if you test in all ticks mode. Even the system of backups is implemented. That is, in case of a server crash, you can transfer trading to another server, you just need to implement automatic copying. This knowledge is enough for me. That is, when I develop an algorithm, I always make sure that the tester matches the demo and real. If it does not match, I look for reasons. As for the fact that it will not reach the real.... I'm not the first day on the market, I've had a lot of things that reached the reals. The last robot traded on real for 2 years, stable, monthly in plus. But! It's all wrong. I want to develop an algorithm that will be at least as reliable as calendar spread strategies.
And on the comments, I still want more criticism like: "it won't work for you because this mechanism can't bring profit, I checked it, look, the maths is against it". So that I can see if I'm wrong somewhere.
The robot is written for real accounts and of course I tested it on demo, the trades fully coincide with the tester, if you test in all ticks mode. Even the system of backups is implemented. That is, in case of server crash, you can transfer trading to another server, you just need to implement automatic copying. This knowledge is enough for me. That is, when I develop an algorithm, I always make sure that the tester matches the demo and real. If it does not match, I look for reasons. As for the fact that it will not reach the real.... I'm not the first day on the market, I've had a lot of things that reached the reals. The last robot traded on real for 2 years, stable, monthly in plus. But! It's all wrong. I want to develop an algorithm that will be at least as reliable as calendar spread strategies.
And on the comments, I still want more criticism like:"it won't work for you because this mechanism can't bring profit, I checked it, look, the maths is against it". So that I can see if I'm wrong somewhere.
the robot is written for real accounts and of course I tested it on demo, the trades fully coincide with the tester, if you test in all ticks mode.
And on the comments, I still want more criticism such as: "it won't work for you because this mechanism can't bring profit, I checked it, look, the maths is against it". So that I can see if I'm wrong somewhere.
Here I will help you to see your mistake.
I didn't read the article, I don't even know what it's about :) but just looked at the graphs and stats for 10 minutes.
I want to ask a few questions first.
1. What server was tested on. If it is on MetaQuotes-Demo server, then there quotes with very low spread and no commission. Sometimes the spread is reduced to zero and even below zero. You can check it.
2. All ticks mode, do you mean "Every tick" mode ? If yes, these are simulated ticks, not real ones.
Try to test in "Every tick based on real ticks" mode.
But all this is nothing.
The main problem you have is the maximum drawdown. If you compare it with monthly growth, there is a very big difference between them. And when you try to increase profitability, you will get Stop Out.
Usually, when brokerage companies or investors are looking for traders to manage their funds, one of the main conditions is this: for 3 months of trading, get a monthly gain of at least 5%, with a total maximum drawdown of no more than 10%.
This is of course not a standard for everyone. But it's roughly like this.
Look at what you are doing.