Stochastic Volatility in Mean: Empirical evidence from Latin-American stock markets using Hamiltonian Monte Carlo and Riemann Manifold HMC methods

Carlos A. Abanto-Valle; Gabriel Rodríguez; Hernán B. Garrafa-Aragón

doi:10.1016/j.qref.2021.02.005

Stochastic Volatility in Mean: Empirical evidence from Latin-American stock markets using Hamiltonian Monte Carlo and Riemann Manifold HMC methods

Carlos A. Abanto-Valle, Gabriel Rodríguez, Hernán B. Garrafa-Aragón

Sección de Economía

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

5 Scopus citations

Abstract

The Stochastic Volatility in Mean (SVM) model of Koopman and Uspensky (2002) is revisited. An empirical study of five Latin American indexes in order to see the impact of the volatility in the mean of the returns is performed. Markov Chain Monte Carlo (MCMC) Hamiltonian dynamics is used to estimate latent volatilities and parameters. Our findings show that volatility has a negative impact on returns, indicating that volatility feedback effect is stronger than the effect related to the expected volatility. This result is clear and opposite to the finding of Koopman and Uspensky (2002).

Original language	English
Pages (from-to)	272-286
Number of pages	15
Journal	Quarterly Review of Economics and Finance
Volume	80
DOIs	https://doi.org/10.1016/j.qref.2021.02.005
State	Published - May 2021

Keywords

Feed-back effect
Hamiltonian Monte Carlo
Markov Chain Monte Carlo
Non linear state space models
Riemannian Manifold Hamiltonian Monte Carlo
Stochastic Volatility in Mean
Stock Latin American markets

Access to Document

10.1016/j.qref.2021.02.005

Cite this

@article{5834106944b24dfc9bd72fd689c55309,

title = "Stochastic Volatility in Mean: Empirical evidence from Latin-American stock markets using Hamiltonian Monte Carlo and Riemann Manifold HMC methods",

abstract = "The Stochastic Volatility in Mean (SVM) model of Koopman and Uspensky (2002) is revisited. An empirical study of five Latin American indexes in order to see the impact of the volatility in the mean of the returns is performed. Markov Chain Monte Carlo (MCMC) Hamiltonian dynamics is used to estimate latent volatilities and parameters. Our findings show that volatility has a negative impact on returns, indicating that volatility feedback effect is stronger than the effect related to the expected volatility. This result is clear and opposite to the finding of Koopman and Uspensky (2002).",

keywords = "Feed-back effect, Hamiltonian Monte Carlo, Markov Chain Monte Carlo, Non linear state space models, Riemannian Manifold Hamiltonian Monte Carlo, Stochastic Volatility in Mean, Stock Latin American markets",

author = "Abanto-Valle, \{Carlos A.\} and Gabriel Rodr{\'i}guez and Garrafa-Arag{\'o}n, \{Hern{\'a}n B.\}",

note = "Publisher Copyright: {\textcopyright} 2021 Board of Trustees of the University of Illinois",

year = "2021",

month = may,

doi = "10.1016/j.qref.2021.02.005",

language = "English",

volume = "80",

pages = "272--286",

journal = "Quarterly Review of Economics and Finance",

issn = "1062-9769",

}

TY - JOUR

T1 - Stochastic Volatility in Mean

T2 - Empirical evidence from Latin-American stock markets using Hamiltonian Monte Carlo and Riemann Manifold HMC methods

AU - Abanto-Valle, Carlos A.

AU - Rodríguez, Gabriel

AU - Garrafa-Aragón, Hernán B.

PY - 2021/5

Y1 - 2021/5

N2 - The Stochastic Volatility in Mean (SVM) model of Koopman and Uspensky (2002) is revisited. An empirical study of five Latin American indexes in order to see the impact of the volatility in the mean of the returns is performed. Markov Chain Monte Carlo (MCMC) Hamiltonian dynamics is used to estimate latent volatilities and parameters. Our findings show that volatility has a negative impact on returns, indicating that volatility feedback effect is stronger than the effect related to the expected volatility. This result is clear and opposite to the finding of Koopman and Uspensky (2002).

AB - The Stochastic Volatility in Mean (SVM) model of Koopman and Uspensky (2002) is revisited. An empirical study of five Latin American indexes in order to see the impact of the volatility in the mean of the returns is performed. Markov Chain Monte Carlo (MCMC) Hamiltonian dynamics is used to estimate latent volatilities and parameters. Our findings show that volatility has a negative impact on returns, indicating that volatility feedback effect is stronger than the effect related to the expected volatility. This result is clear and opposite to the finding of Koopman and Uspensky (2002).

KW - Feed-back effect

KW - Hamiltonian Monte Carlo

KW - Markov Chain Monte Carlo

KW - Non linear state space models

KW - Riemannian Manifold Hamiltonian Monte Carlo

KW - Stochastic Volatility in Mean

KW - Stock Latin American markets

UR - https://www.scopus.com/pages/publications/85101975626

U2 - 10.1016/j.qref.2021.02.005

DO - 10.1016/j.qref.2021.02.005

M3 - Article

AN - SCOPUS:85101975626

SN - 1062-9769

VL - 80

SP - 272

EP - 286

JO - Quarterly Review of Economics and Finance

JF - Quarterly Review of Economics and Finance

ER -

Stochastic Volatility in Mean: Empirical evidence from Latin-American stock markets using Hamiltonian Monte Carlo and Riemann Manifold HMC methods

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this