Simulation-based sequential analysis of Markov switching stochastic volatility models

Carlos M. Carvalho; Hedibert F. Lopes

doi:10.1016/j.csda.2006.07.019

Simulation-based sequential analysis of Markov switching stochastic volatility models

Carlos M. Carvalho
, Hedibert F. Lopes

Research output: Contribution to journal › Article › peer-review

76 Scopus citations

Abstract

We propose a simulation-based algorithm for inference in stochastic volatility models with possible regime switching in which the regime state is governed by a first-order Markov process. Using auxiliary particle filters we developed a strategy to sequentially learn about states and parameters of the model. The methodology is tested against a synthetic time series and validated with a real financial time series: the IBOVESPA stock index (São Paulo Stock Exchange).

Original language	English (US)
Pages (from-to)	4526-4542
Number of pages	17
Journal	Computational Statistics and Data Analysis
Volume	51
Issue number	9
DOIs	https://doi.org/10.1016/j.csda.2006.07.019
State	Published - May 15 2007
Externally published	Yes

Keywords

Bayes factor
Bayesian time series
Markov chain Monte Carlo
Particle filters
Sequential analysis
Stochastic volatility models

ASJC Scopus subject areas

Statistics and Probability
Computational Mathematics
Computational Theory and Mathematics
Applied Mathematics

Access to Document

10.1016/j.csda.2006.07.019

Cite this

@article{3249d3ef66a34eab865808e465bca83e,

title = "Simulation-based sequential analysis of Markov switching stochastic volatility models",

abstract = "We propose a simulation-based algorithm for inference in stochastic volatility models with possible regime switching in which the regime state is governed by a first-order Markov process. Using auxiliary particle filters we developed a strategy to sequentially learn about states and parameters of the model. The methodology is tested against a synthetic time series and validated with a real financial time series: the IBOVESPA stock index (S{\~a}o Paulo Stock Exchange).",

keywords = "Bayes factor, Bayesian time series, Markov chain Monte Carlo, Particle filters, Sequential analysis, Stochastic volatility models",

author = "Carvalho, \{Carlos M.\} and Lopes, \{Hedibert F.\}",

note = "Funding Information: The results presented here were developed in collaboration with the author{\textquoteright}s Ph.D. students Jason Williams and Brian Whited (now at Disney Animation), and with colleagues Fr{\'e}d{\'e}ric Chazal (INRIA), Andr{\'e} Lieutier, (Dassault Systmes), Greg Slabaugh (formerly at Siemens), and Tong Fang and Gozde Unal (both from Siemens). Nick Patrikalakis (MIT), Chris Hoffman (Purdue), and Bruno Levy and Nicolas Ray (INRIA) have also provided valuable input to some aspects of this work.",

year = "2007",

month = may,

day = "15",

doi = "10.1016/j.csda.2006.07.019",

language = "English (US)",

volume = "51",

pages = "4526--4542",

journal = "Computational Statistics and Data Analysis",

issn = "0167-9473",

publisher = "Elsevier B.V.",

number = "9",

}

TY - JOUR

T1 - Simulation-based sequential analysis of Markov switching stochastic volatility models

AU - Carvalho, Carlos M.

AU - Lopes, Hedibert F.

N1 - Funding Information: The results presented here were developed in collaboration with the author’s Ph.D. students Jason Williams and Brian Whited (now at Disney Animation), and with colleagues Frédéric Chazal (INRIA), André Lieutier, (Dassault Systmes), Greg Slabaugh (formerly at Siemens), and Tong Fang and Gozde Unal (both from Siemens). Nick Patrikalakis (MIT), Chris Hoffman (Purdue), and Bruno Levy and Nicolas Ray (INRIA) have also provided valuable input to some aspects of this work.

PY - 2007/5/15

Y1 - 2007/5/15

N2 - We propose a simulation-based algorithm for inference in stochastic volatility models with possible regime switching in which the regime state is governed by a first-order Markov process. Using auxiliary particle filters we developed a strategy to sequentially learn about states and parameters of the model. The methodology is tested against a synthetic time series and validated with a real financial time series: the IBOVESPA stock index (São Paulo Stock Exchange).

AB - We propose a simulation-based algorithm for inference in stochastic volatility models with possible regime switching in which the regime state is governed by a first-order Markov process. Using auxiliary particle filters we developed a strategy to sequentially learn about states and parameters of the model. The methodology is tested against a synthetic time series and validated with a real financial time series: the IBOVESPA stock index (São Paulo Stock Exchange).

KW - Bayes factor

KW - Bayesian time series

KW - Markov chain Monte Carlo

KW - Particle filters

KW - Sequential analysis

KW - Stochastic volatility models

UR - https://www.scopus.com/pages/publications/34147112845

UR - https://www.scopus.com/pages/publications/34147112845#tab=citedBy

U2 - 10.1016/j.csda.2006.07.019

DO - 10.1016/j.csda.2006.07.019

M3 - Article

SN - 0167-9473

VL - 51

SP - 4526

EP - 4542

JO - Computational Statistics and Data Analysis

JF - Computational Statistics and Data Analysis

IS - 9

ER -

Simulation-based sequential analysis of Markov switching stochastic volatility models

Abstract

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this