Top Traders Unplugged
SI266: Constructing and Assessing Trading Strategies ft. Rob Carver
Oct 21, 2023
Rob Carver, an expert in trading strategies, discusses trend following and mean reversion strategies, the benefits of faster trend following, and avoiding over-fitting when constructing trading strategies. They also explore the performance of a diversified program versus a more concentrated one and the correlation between trend trading and equities.
01:02:28
Episode guests
AI Summary
AI Chapters
Episode notes
Podcast summary created with Snipd AI
Quick takeaways
- To prevent overfitting, it is important to strike a balance between fitting the data and maximizing out-of-sample performance.
- Quantifying fitting and robustness can be challenging, but techniques such as examining performance across different time periods and using bootstrapping can help assess a strategy's characteristics.
Deep dives
Detecting Overfitting and Maximizing Out-of-Sample Performance
Overfitting occurs when a strategy is too closely aligned with past data and assumes the future will be similar. To prevent overfitting, it is important to strike a balance between fitting the data and maximizing out-of-sample performance. This means capturing the broad essence of the data without excessively tailoring the strategy to specific instances. One way to detect overfitting is by examining the performance of the strategy across different sub-samples or alternative histories. By analyzing the distribution of performance, we can assess whether the strategy is robust and performs well across various scenarios. The goal is to have a strategy that performs consistently well across different time periods or samples.
Remember Everything You Learn from Podcasts
Save insights instantly, chat with episodes, and build lasting knowledge - all powered by AI.
