DiT-LSTM-SVAR Model For Portfolios
Keywords: DiT-LSTM, SVAR, Portfolios
TL;DR: DiT-LSTM-SVAR Model For Portfolios
Abstract: This paper proposes a novel combined model named DiT-LSTM-SVAR, which successfully integrates time series and the Efficient Markets Hypothesis. This is the first to combine the microstructure of financial markets with deep learning networks to improve the performance of portfolios. We employ the DiT model to predict the upside and downside movements and an information decomposition model based on the SVAR model to identify random walk stocks. The DiT module significantly improves the Matthews correlation coefficient by almost 3\%. The annual return of the portfolio is improved by almost 20\%. The SVAR module greatly improves the Matthews correlation coefficient by almost 4\%. Portfolios constructed using the DiT-LSTM-SVAR module based on market and public information outperformed those created with the DiT-LSTM model. The annual cumulative return of the portfolio is 266.60\% and a Sharpe ratio of 1.8.
Primary Area: other topics in machine learning (i.e., none of the above)
Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics.
Submission Guidelines: I certify that this submission complies with the submission instructions as described on https://iclr.cc/Conferences/2025/AuthorGuide.
Reciprocal Reviewing: I understand the reciprocal reviewing requirement as described on https://iclr.cc/Conferences/2025/CallForPapers. If none of the authors are registered as a reviewer, it may result in a desk rejection at the discretion of the program chairs. To request an exception, please complete this form at https://forms.gle/Huojr6VjkFxiQsUp6.
Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors’ identity.
No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.
Submission Number: 5367
Loading