MIS - Automatic Generated Statistic
Concept diagram
The concept diagram provided below depicts the function block associated with this page.
Signal statistic generation
The Expert Advisor Builder can generate comprehensive statistics that encapsulate vital information about each signal generator. This feature is exclusively accessible when creating a trading module with the Expert Advisor Builder. However, when the trading module is later executed with the Custom Expert Advisor, the statistic generator is deactivated for performance optimization.
If you wish to activate the statistic generator, please consult the input parameter below.
| Input variable within MetaTrader (Identifier in SET file) | Default value | Comment |
|---|---|---|
| > (MIS) Statistic generator (CPU and HDD intensive!) (IN_enmMisGenerateStatisticOutput) | OFF | Numeric values usable in SET files: 0 - OFF 1 - ON |
The Expert Advisor Builder will populate its Statistic output folder with data upon activation. As illustrated in the subsequent folder structure, you can access the statistics folder via the tester data folder. Delving deeper into the structure, you'll find the Final and Raw folders and an Excel file named Statistic Evaluation. This Excel file serves to import and display the generated statistical data. For a more in-depth understanding of the file's usage and diagrams, please refer to the corresponding article in the user manual.
The Final and Raw folders house CSV files that can be imported and analyzed using the Statistical Evaluation Excel file. However, if you prefer to process the information computationally, you can use the CSV files in the Final folder directly.
Value Analysis Diagrams
The Value Analysis Diagram sheet within the evaluation Excel file contains diagrams depicting each signal value's frequency during the tested period. As indicated in the image below, a diagram is available for each signal-generating functional block.
Let's examine the statistics for the timing module. The statistic generator produces data reflecting the typical rate of change per hour, day of the week, day of the month, and week of the year. Here, the rate of change is determined by the difference between the highest and lowest values during the measured period relative to the price at the end of that period. For instance, the Day of Week diagram suggests that the rate of change is most significant on Thursdays, indicating that the observed market was statistically more volatile on Thursdays and least volatile on Mondays.
The subsequent diagram presents the angle and value of the INB2 indicator block over the duration of the trading module's execution in the strategy tester. The most frequent angles are above zero, suggesting that the market was bearish for most of the observed period. Particularly for indicators, the value diagram can highlight support and resistance values.
The statistical data provides diverse perspectives on the signals generated by a trading module. This information can be used to adjust the signal input parameters appropriately or serve as input for automated parameter adaptation.
Signal Analysis Diagrams
The Signal Analysis Diagram sheet in the evaluation Excel file includes diagrams that depict the average chart shape before and after each signal occurrence. Each signal-generating block is represented by a unique diagram. Additionally, the actual trading signal diagram only uses the signal for opening trades, while the general signal means the combination of all signal generators. For a more detailed explanation of these statistical diagrams, please refer to the specific article in the user manual.
Generally, each diagram consists of numerous rectangles, each representing the rate of change in the bars before and after the signal in the timeframe in which the trading module is executed. Each signal's rate of change values serve as input to calculate the average bar range and position. For instance, the following diagram displays the average chart shape before and after each trading signal that opens a trade. As can be inferred, the chart predominantly declined before the trades were opened, but the remaining shape indicated sideways-moving markets while the trades were open. A trading strategy resulting in a chart like this is unlikely to yield long-term profits.
This type of diagram is produced for every signal-generating function block. The subsequent image shows the average chart shape before and after the signals generated by the chart analyzer CA1. This generally appears promising for executing short trades since the average shape decreases after the signal moment.
The signal analysis data is a powerful resource for automated input configuration with an external script. Please note that this topic is beyond the scope of this user manual.