What is it about?
We create a sequentially optimised ranking algorithm that selects cross-sectional momentum (long/short) sub-portfolios whose constituents are chosen from the S&P 500 index. We outperform the long-only holding of the index with hindsight and a regress-then-rank baseline.
Featured Image
Photo by vackground.com on Unsplash
Why is it important?
Unlike earlier algorithms, our ranker can operate on nonstationary time series or those that experience regular covariate shifts or concept drifts. If the best experts change over time, our ranker will adapt and find them.
Read the Original
This page is a summary of: Sequential asset ranking in nonstationary time series, October 2022, ACM (Association for Computing Machinery),
DOI: 10.1145/3533271.3561666.
You can read the full text:
Contributors
The following have contributed to this page
