It is common that a trading algorithm involves a number of user defined trading parameters. To optimize the algorithm's trading performance, developers usually need to tweet the parameters, re-run backtest, and compare the results manually.
ALGOGENE is excited to announce a new system feature to support such parameter optimization process!
Before going into the Strategy Optimization function, all we need to do is to develop a backtest script as usual. Let's borrow the code from this article [A simple RSI strategy].
- First of all, we need to backtest our strategy script to create a "finished" backtest record. It is to ensure that our base strategy is bug-free.
- Go to [My History] > [Finished Backtest], under your base strategy, click "Optimize"
- It will load our base strategy in left panel, we then need to set the following, and click "Optimize"
- Optimization objective
- Optimization method
- The parameters set to optimize
- from a starting value to the end value inclusively, with an incremental step size
- 'Value' specifies the initial estimate for the optimization problem
- It will take some time for the engine to generate the backtest results. Depending on the complexity of our base strategy and the grid size of our parameter sets, the overall optimization process may ranging from a few minutes to a couple of days.
- Upon the optimization process finished, we can obtain the
- performance summary
- visualization charts
- cummulative returns
Load optimization results
When we reload [My History] page, the optimization results can be found under the same group ID. We can load to see the backtest result, or compare with other optimization results.
Now, we got a new tool for speed up the algo development process! Happy Trading!