Estimation and application of semiparametric stochastic volatility models based on kernel density estimation and hidden Markov models. (19th January 2018)

• Record Type: Journal Article
• Authors: Hao, Hong‐Xia; Lin, Jin‐Guan; Huang, Xing‐Fang; Wang, Hong‐Xia; Zhao, Yan‐Yong
• Other Names: Nicolis Orietta guestEditor.; Pascual Rodrigo guestEditor.
• Is Part Of: Applied stochastic models in business and industry. Volume 34:Number 3(2018)
• Page Start: 355
• Page End: 375
• Publication Date: 2018-01-19
• Subjects: forward algorithm -- hidden Markov model -- kernel density estimation -- numerical integration -- stochastic volatility models; Stochastic analysis -- Periodicals; Stochastic processes -- Periodicals; Business mathematics -- Periodicals; Finance -- Mathematical models -- Periodicals; Industrial management -- Mathematical models -- Periodicals; 338.00151923
• DOI: 10.1002/asmb.2305
• Languages: English
• ISSNs: 1524-1904

Abstract

Abstract : Discrete‐time stochastic volatility models play a key role in the analysis of financial time series. However, the parametric assumption of conditional distribution for asset returns, given the volatility, has been questioned. When the conditional distribution is unknown and unspecified, in this paper, a maximum‐likelihood estimation approach for the semiparametric stochastic volatility models is proposed based on kernel density estimation and hidden Markov models. Several numerical studies are conducted to evaluate the finite sample performance of the proposed estimation method. Implementation on empirical studies also illustrates the validity of the proposed method in practice.