In order to describe the algorithm for working with the system, it is first necessary to indicate the connections that exist between the blocks of the system. The user interaction block is associated with the data acquisition block, because the user specifies the stock and the time period on which the models will be built and trained. Also, the user interaction block is associated with the data visualization block, since it is the user who initiates the display of a particular graph. Another block of user interaction is associated with the trading block, because it is the user who, evaluating the results of the analysis, decides whether to conduct trading simulation or not. The data acquisition block is connected to two other blocks, namely, the database operation block and the data analysis block. The connection with the database block is due to the fact that the data downloaded from the site must be placed in the database in order to maintain historical records of the tests performed. Communication with the data analysis block is due to the fact that the data acquisition block prepares data for analysis, and then transfers them to the data analysis block.
The data analysis block is only directly related to the test block. After the models are built, data about these models is passed to the test block for testing and checking the quality of the models. The test block is associated with two blocks: the block for working with the database and the block for interacting with the user. In the case of the first relationship after testing and calculating the performance criteria, the data is processed and stored in a database for storage. In the case of the user interaction block, the results of the test run of data and the calculation of criteria are published in the application, where the user is already studying them. Having analyzed all the interconnections of the system, it is possible to describe the algorithm of the system operation (see Figure 14). The system starts from the moment when the user enters data and sends it for further analysis.
The query criteria is then passed from the user interaction block to the data retrieval block. After the download and processing of the data is completed, they are stored in the database and simultaneously transferred to the data analysis unit. Based on the downloaded and prepared data, models are built and then information about all the built models is transferred to the test block. In the test block, all models are tested on some test datasets. Based on the results obtained, the efficiency criteria are calculated. Then all this data is stored in the database and displayed to the user. After receiving the results of the analysis, the user can display the data on a chart or run a simulated trade.
This completes the cycle of the system, and the user can initiate a reanalysis for another stock or for another time period. .