Who wants a strategy? Lots and for free) - page 25
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
The repository has been up and running for a long time. Only the control of it is built into another program. Now I'm transferring the code to SSB.
And the technology has already been described in the previous posts of this branch.
In my opinion, you described the principles, not the technology. And it, imho, is just about where, in which program, on which server (and/or user's computer) the processing is done, where and how the results are stored, etc. That's why I asked for details.
In my opinion, you described the principles, not the technology. And it is, imho, exactly where, in which software, on which server (and/or user's computer) the processing is done, where and how the results are stored, etc.
PHP + MySQL on the server ssb.bigfx.ru
A repository-enabled version of SSB is ready and posted. Those who would like to try it can follow this link:
Download Stock Strategy Builder version 3.0
The essence of the repository is that having generated one strategy, you can get from the repository via the Internet several more strategies for the same symbol and timeframe (but no more than 20), having the maximum rating. The rating of each strategy depends on the results of its testing, which is carried out immediately after its downloading from the repository.
There are not so many strategies in the repository at the moment. However, as the number of repository users increases, the number of strategies in the database should also increase.
The repository-enabled version of SSB is ready and posted. Those who would like to try it can follow this link:
Download Stock Strategy Builder version 3.0
The essence of the repository is that having generated one strategy, you can get from the repository via the Internet several more strategies for the same symbol and timeframe (but no more than 20), having the maximum rating. The rating of each strategy depends on the results of its testing, which is carried out immediately after its downloading from the repository.
There are not many strategies in the repository at the moment. But with increasing number of repository users, the number of strategies in the database should also increase.
Thank you very much, Yuriy!
Today I ran the first statistics on the repository base:
Strategies with a positive rating (i.e. above 0): 9.52380952380239523802395238095%
Strategies with zero and negative ratings: -90.4761904747619047619047619%.
That is, less than 10% of strategies that were showing good results in individual tests, after a more comprehensive testing by repository users, turned out to be among outsiders - scams. And these results were obtained in less than a day and on a negligibly small base.
It is clear that as time passes, the percentage of strategies with the positive rating will decrease, while the percentage of scams will increase.
I hope you understand now why the repository was created. If we loop the process of searching for strategies and complex checking of strategies on one computer, it will take a very significant time to get more or less suitable results for analysis and conclusions. But a repository based on distributed computing can weed out all sorts of scum in less time and with much less computational effort.
I don't understand the sequence of operation, if you can be more specific.
1,I start the program SSB.
2, The terminal opens and starts optimization, selection of strategies.
3.After completion of selection terminal closes and starts visualization, increasing the speed of visualization to the maximum terminal closes and everything starts over again, nothing and no results the program does not show.
I don't understand the sequence of operation, if you can be more specific.
1,I start the program SSB.
2, I open the terminal and start the optimization and strategy selection.
3. After completing the selection terminal closes and starts visualization, increasing the speed of visualization to the maximum terminal closes and everything starts anew, nothing and no results progazdaetsya.
Visualization should be turned off. And do this in the tester before starting the SSB
If the Internet is connected, then the SSB:
1. It selects the most acceptable strategy by parameters
2. Connects to the repository and downloads the list of strategies with the maximum rating
3. As strategies are downloaded, it runs tests and sends the results of these tests to the repository
4. Saves the results of the tests in a local list
5. When the download of strategies from the repository is completed, a message is displayed stating that the download has been completed successfully and the internet can be disconnected, if it is a pay-per-view.
6. Starts running the strategies saved in the list and returns the results and offers to save them in the MQL4 code, if the result is satisfying
The process is not very fast, so if the visualisation is switched on, then it will take a lot of time to test each strategy downloaded.
When the internet is switched off, items 2, 3 and 5 are not executed. 2, 3 and 5 are not performed.
5. Upon completion of the download of strategies from the repository, a message is displayed stating that the download has been completed successfully and the internet can be disconnected if the payment is for a fee.
In the repository, strategies are sifted by currency pairs and by timeframe. If the strategy was made on the Euradollar, but tested on the pound then let's say it will automatically lower the rating of the strategy.
Is it supposed to reboot the terminal after optimisation?
The terminal cannot be started from the command line by an external program without rebooting. That's why it works like this.
And in which directory it saves the selected strategies?
Strategies are in the same folder as ssb_3.exe
Impeller писал(а) >>
In the repository, strategies are sifted by currency pairs and by timeframes. If the strategy was done on EURUSD and tested on GBPUSD, it will automatically decrease the strategy rating.No, there is no division at the moment. Later, when the database is full, I will filter by timeframe only.
Many strategies are suitable for most financial instruments. I am watching the ratings now. Leading strategies have only increased rating, outsiders have only decreased rating. And fresh strategies gradually move to one side or the other. Rarely do they linger near the initial zero rating.