This e-book contains a description of the workflow that I use for profit when trading via genetic algorithms (specifically using StrategyQuant platform). This workflow consists of several important parts. I am briefly introducing them and in ebook I am going to tell you about them in more detail:
Portfolio of strategies: What we are looking for
It is important to realize what we want to achieve with our strategies. We will discuss it in the first part.
Import of data
The first important step is to import data correctly and set them accordingly. This is the most important step, because the result of backtests and robustness tests depends on the quality of input data. If the quality of the data is poor, the result of the backtest will also be poor and not reliable. Spend time on this part and learn how to process the data correctly.
Generating an initial package of strategies
Once the data is set up, it is necessary to generate an initial package of strategies. Ideally we build 1000 to 2000 strategies and then we do all the robustness tests and select those that are most likely to be profitable in the future.
Robustness tests
This is the crucial part. The robustness tests let you determine which strategies are most likely to be profitable in real trading and also how vulnerable they are to possible changes, such as a change of volatility, slippage, etc.
Walk forward matrix
Testing that uses the walk forward matrix method determines how reliable the strategy is and how much can be expected from it in the future. This is the most reliable test available.
