Tail-Driven Nonparametric Estimation for State Price Densities

• 成果类型: Article; Early Access
• 署名作者: Li, Chenxu; Song, Xiaojun; Wan, Yating
• 署名单位: Peking University; Tianjin University of Finance & Economics
• 刊物名称: MANAGEMENT SCIENCE
• ISSN/ISSBN: 0025-1909
• DOI: 10.1287/mnsc.2023.03236
• 发表日期: 2026
• 关键词: state price density; Nonparametric Estimation; Tail risk; risk management; asset pricing

摘要

This paper proposes and implements a novel nonparametric method for estimating the state price density (SPD) over the entire state space, including the tails. This SPD estimator achieves shape consistency properties in theory, particularly at the tails. Monte Carlo simulations demonstrate the accuracy and robustness of our method. In particular, our estimator accurately captures the risk-neutral tail distribution, which is often underestimated by existing alternative methods. In an empirical analysis based on Standard and Poor's 500 options data, we evaluate the out-of-sample performance of our SPD estimation method and demonstrate that the estimates can serve as effective indicators for market conditions and exhibit predictive power for asset returns. Combining these perspectives, we suggest that our SPD estimator renders a valuable tool for risk management and asset pricing.