Do Better Volatility Forecasts Lead to Better Portfolios? Evidence from Graph Neural Networks
For quantitative finance practitioners, this paper provides empirical evidence that forecast accuracy, ranking quality, and portfolio performance are distinct objectives, challenging the assumption that better forecasts directly lead to better portfolios.
The paper tests whether graph neural networks improve volatility forecasts and portfolio performance for 465 S&P 500 equities (2015–2025). It finds that the best forecasting model, best ranking model, and best portfolio model are different, and graph models add value only when the portfolio rule exploits cross-sectional structure.
This paper tests whether graph neural networks improve realized volatility forecasts and whether those forecasts improve portfolio performance. Using weekly realized volatility for 465 S\&P 500 equities from 2015--2025, Heterogeneous Autoregressive and Long Short-Term Memory baselines are compared against GraphSAGE models built on rolling correlation, sector, and Granger-causal graphs, with and without macro regime features. The empirical finding is that the model with the lowest forecast MSE, the model with the highest cross-sectional ranking accuracy, and the model with the highest portfolio Sharpe ratio are three different models. Forecast accuracy, ranking quality, and portfolio performance are related but not interchangeable objectives. Graph volatility models add value only when the portfolio rule can exploit the cross-sectional structure they encode.