Financially Guided Deep Portfolio Optimization
For quantitative finance practitioners, this work demonstrates that embedding financial objectives directly into neural network training can produce robust portfolio allocations that outperform traditional methods and benchmarks, even in adverse market conditions.
The paper proposes an end-to-end deep portfolio optimization framework that directly optimizes differentiable surrogates of financial metrics (Sharpe ratio, Omega ratio, CVaR, Risk Parity). On 50 S&P 500 stocks from 2007-2023, the best model (AttentionLSTM with Omega-CVaR-RiskParity loss) achieves an annualized Sharpe of 0.29 and +7.86% total return in 2022-2023, outperforming the S&P 500 by 12.38 percentage points.
Portfolio optimization in real-world financial markets is notoriously difficult due to non-stationarity, noisy data, and high transaction costs. Standard predict-then-optimize methods first forecast returns and then solve for weights, compounding prediction errors and often failing under regime shifts. We propose an end-to-end framework that directly optimizes differentiable surrogates of key financial metrics - Sharpe ratio, Omega ratio, Conditional Value-at-Risk (CVaR), and Risk Parity - allowing neural networks to learn portfolio weights via backpropagation. Our expanding-window walk-forward procedure, applied to 50 S&P 500 stocks from 2007 to 2023, incorporates realistic bid-ask spread costs and rebalances quarterly. On the challenging out-of-sample test period (2022-2023), the best model - an AttentionLSTM with the Omega-CVaR-RiskParity loss - achieves an annualized Sharpe of 0.29 and a total compounded return of +7.86%, while the S&P 500 delivers -4.52% total return and an annualized Sharpe of -0.02. This outperforms the S&P 500 by 12.38 percentage points (a relative improvement of over 270%), while keeping tail risk (CVaR) nearly unchanged. The framework consistently outperforms the equal-weight portfolio, S&P 500, and traditional methods (MVP, HRP, NCO), demonstrating that embedding financial objectives directly into model training yields robust, economically meaningful outperformance even in adverse market conditions.