Discussing the article: "Developing a multi-currency Expert Advisor (Part 9): Collecting optimization results for single trading strategy instances"
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
Check out the new article: Developing a multi-currency Expert Advisor (Part 9): Collecting optimization results for single trading strategy instances.
Let's outline the main stages of the EA development. One of the first things to be done will be to optimize a single instance of the developed trading strategy. Let's try to collect all the necessary information about the tester passes during the optimization in one place.
Essentially, the main type of data we need to store and use is the optimization results of multiple EAs. As you know, the strategy tester records all optimization results in a separate cache file with the *.opt extension, which can then be reopened in the tester or even opened in the tester of another MetaTrader 5 terminal. The file name is determined from the hash calculated based on the name of the optimized EA and the optimization parameters. This allows us not to lose information about the passes already made when continuing optimization after its early interruption or after changing the optimization criterion.
Therefore, one of the options under consideration is the use of optimization cache files to store intermediate results. There is a good library from fxsaber allowing us to access all saved information from MQL5 programs.
But as the number of optimizations performed increases, the number of files with their results will also increase. In order not to get confused, we will need to come up with some additional structure for arranging the storage and working with these cache files. If optimization is not carried out on one server, then it will be necessary to implement synchronization or storing all cache files in one place. In addition, for the next stage we will still need some processing to export the obtained optimization results to the EA at the next stage.
Then let's look at arranging the storage of all results in the database. At first glance, this would require quite a lot of time to implement. But this work can be broken down into smaller stages, and we will be able to use its results immediately, without waiting for full implementation. This approach also allows for greater freedom in choosing the most convenient means of intermediate processing of stored results. For example, we can assign some processing to simple SQL queries, something will be calculated in MQL5, and something in Python or R programs. We will be able to try different processing options and choose the most suitable one.
Author: Yuriy Bykov