Maximum likelihood estimation of latent affine processes

• 成果类型: Article
• 署名作者: Bates, David S.
• 署名单位: University of Iowa
• 刊物名称: REVIEW OF FINANCIAL STUDIES
• ISSN/ISSBN: 0893-9454
• DOI: 10.1093/rfs/hhj022
• 发表日期: 2006
• 页码: 909
• 关键词: EMPIRICAL CHARACTERISTIC FUNCTION; STOCHASTIC VOLATILITY MODEL; MOMENTS ESTIMATION; simulated moments; GMM estimation; term structure; options; EFFICIENCY; variance; implicit

摘要

This article develops a direct filtration-based maximum likelihood methodology for estimating the parameters and realizations of latent affine processes. Filtration is conducted in the transform space of characteristic functions, using a version of Bayes' rule for recursively updating the joint characteristic function of latent variables and the data conditional upon past data. An application to daily stock market returns over 1953-1996 reveals substantial divergences from estimates based on the Efficient Methods of Moments (EMM) methodology; in particular, more substantial and time-varying jump risk. The implications for pricing stock index options are examined.