Automated Trading Strategy 75: Development Tips From The AI 'AG'
Update On AI Generative Strategy Performance & 9 More Strategies Generated by the AG
Important: There is no guarantee that these strategies will have the same performance in the future. I use backtests to compare historical strategy performance. Backtests are based on historical data, not real-time data so the results shared are hypothetical, not real. There are no guarantees that this performance will continue in the future. Trading futures is extremely risky. If you trade futures live, be prepared to lose your entire account. I recommend using these strategies in simulated trading until you/we find the holy grail of trade strategy.
As a quick reminder, our goal is to find the holy grail of automated trade strategy.
We haven’t found the holy grail yet, but we get closer with every strategy. Click here for the most recent performance chart and links to all strategies. For answers to some FAQs, click here.
Contact: AutomatedTradingStrategies@protonmail.com.
This is a follow up to the post Strategy 70 which originally introduced Ninjatrader’s AI Generate (aka ‘AG’) as a generative AI tool. For a full review of how ATS is using AI, please read the original post.
I’ve decided to split this strategy into two parts because there’s a lot to digest. Part 2 will be posted in a day or so and will introduce the use of features we’ve never discussed before such as crossover rate and mutuation stregnth.
First, let’s start with a quick review of AG.
Strategy Development Tips From The Mind of The AI Called AG
AG does many things, but at a high level it tries to find the best combination of indicators and candlestick formations given a certain time series. You can also drill down based on session time, day of week, etc., and use multiple stop/profit options like Parabolic stops, Trailing Stops, Profit Targets etc. It works by applying a genetic algorithm to a group of predefined indicators and patterns. We go from evaluating every possible solution to an evolutionary approach that continuously improves the solution set. In this way, the genetic algorithm can explore larger data sets by searching for evolutionary ‘promising’ regions rather than systematically searching through all the data in brute force fashion.
In July of 2023, I published 7 strategies that were all created by the AI AG. Today, I want to go back and look at how well those strategies are performing. We’ll also ‘open the kimono’ to unveil the code behind the best performing strategy. What I found was something quite different from anything we’ve used in our strategies and I’m looking forward to sharing these new ideas with you.
I’m also going to provide 9 additional strategies created by AI AG for download. If you’ve used AG you know that finding 9 profitable strategies is not an easy thing to do. We’ll be adding these strategies to the forward test as well.
In a day or so I’ll publish Part 2 of Strategy 75. I thought it prudent to break this post into two parts since AI AG is not the same as the optimization model that uses the Genetic Algorithm—the former is a generative algo that generates new strategies, the latter uses the genetic algo to help find the best optimization. We’ll do a deep dive into its feature set, especially the mutation function (my favorite).
Now, let’s get into it.
AG’s Performance Results
The chart below provides the forward test start date, trade count and net income of AGs strategies as published by ATS in Strategy 70: