Portfolio optimization using deep learning with risk aversion utility function

Kenji Kubo and Kei Nakagawa

Finance Research Letters, 2025, vol. 74, issue C

Abstract: This paper explores portfolio optimization with deep learning (DL), which can model non-linear returns that traditional methods cannot capture. While Sharpe loss addresses the risk-return trade-off in DL-based portfolio construction, it has limitations, including interpretability issues with negative PnL and biased gradients under stochastic gradient descent (SGD). We propose a new loss function based on a risk-averse utility function, which provides unbiased gradients and clear interpretation even with negative PnL. Additionally, we use DL outputs as adjustments to baseline weights, achieving improved portfolio performance. Experiments on S&P 500 data show that our method outperforms Sharpe loss-based models across several metrics, including the Sharpe ratio.

Keywords: Portfolio optimization; Deep learning; Risk-averse utility; Sharpe loss (search for similar items in EconPapers)
Date: 2025
References: Add references at CitEc
Citations:

Downloads: (external link)
http://www.sciencedirect.com/science/article/pii/S1544612325000261
Full text for ScienceDirect subscribers only

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:eee:finlet:v:74:y:2025:i:c:s1544612325000261

DOI: 10.1016/j.frl.2025.106761

Access Statistics for this article

Finance Research Letters is currently edited by R. Gençay

More articles in Finance Research Letters from Elsevier
Bibliographic data for series maintained by Catherine Liu ().