A STUDY ON THE RETURN VOLATILITY OF THE COFFEE BEANS PRICE USING ARCH MODELS
Keywords:
Bayesian approach, Informative prior distributions, MCMC methodsAbstract
In the case of primary commodities, price volatility would arise mainly due to disturbances in supply, whereas for industrial raw materials, it would be the result of disturbances in demand. In the analysis of commodity markets can be seen that information, hedging, speculation and physical availability are factors that can influence their volatility. Moreover, increased volatility in commodity markets can justify the use of information-based processes for modeling the pattern of return volatility of these commodities. Since the relevance of autoregressive conditional heteroscedasticity (ARCH) family models in the solution of problems in economic and financial areas due to their applicability and interpretation (the relation between return and volatility) have been provided, the aim of this work was to compare the Bayesian estimates for the parameters of ARCH processes with normal and Student’s t distributions for the conditional distribution of the return series of coffee beans price. In addition, informative prior distributions were suggested and posterior summaries were obtained by Monte Carlo Markov Chain simulation methods. Results showed that the proposed Bayesian approach provides satisfactory estimates and that the ARCH process with Student’s t distribution adjusts better to the data.
Downloads
References
ANDRADE, M. G.; OLIVEIRA, S. C. A comparative study of Bayesian and Maximum Likelihood approaches for ARCH models with evidence from Brazilian financial series. New Mathematics and Natural Computation, v.7, n.2, p. 347-361, 2011. https://doi.org/10.1142/S1793005711001974
AUSÍN, M. C.; GALEANO, P. Bayesian Estimation of the Gaussian Mixture GARCH Model. Computational Statistics and Data Analysis, v.51, n.5, p.2636-2652, 2007. https://doi.org/10.1016/j.csda.2006.01.006
BARRETO, G. A.; OLIVEIRA, S. C.; ANDRADE, M.G. Estimação de Parâmetros de Modelos ARCH(p): Abordagem Bayes-MCMC versus Máxima Verossimilhança. Revista Brasileira de Estatística, v.69, n.1, p.7-24, 2008.
BOLLERSLEV, T. Generalized Autoregressive Conditional Heteroscedasticity. Journal of Econometrics, v.31, n.3, p.307-327, 1986. https://doi.org/10.1016/0304-4076(86)90063-1
BOLLERSLEV, T. Glossary to ARCH (GARCH). Research paper, School of Economics and Manangement, University of Aarthus (2008). Electronic copy available at: http://ssrn.com/abstract=1263250. https://doi.org/10.2139/ssrn.1263250
CARLIN, B.; CHIB, S. Bayesian Model Choice via Markov Chain Monte Carlo Methods. Journal of the Royal Statistical Society, v.57, n.3, p.473-484, 1995. https://doi.org/10.1111/j.2517-6161.1995.tb02042.x
CASELLA, G. E.; GEORGE, E. I. Explaining the Gibbs sampler. The American Statistician, v.46, n.8, p.167-174, 1992. https://doi.org/10.1080/00031305.1992.10475878
CHIB, S.; GREENBERG, E. Understanding the Metropolis-Hastings Algorithm. The American Statistician, v.49, n.4, p.327-335, 1995. https://doi.org/10.1080/00031305.1995.10476177
ENDERS, W. Applied Econometric Times Series. New Jersey: Wiley, 2009. 544p.
ENGLE, R. Autoregressive Conditional Heteroscedasticity with Estimates of the Variance of UK Inflation. Econometrica, v.50, n.4, p.987-1007, 1982. https://doi.org/10.2307/1912773
GEWEKE, J. Exact Predictive Densities for Linear Models with ARCH Disturbances. Journal of Econometrics, v.40, n.1, p. 63-86, 1989. https://doi.org/10.1016/0304-4076(89)90030-4
GEWEKE, J. Evaluating the Accuracy of sampling-based approaches to calculating posterior moments (with discussion). Bayesian Statistics, v.4, p.164-193, 1992.
MORETTIN, P. A. Econometria Financeira. São Paulo: Edgard Blucher Ltd., 2008. 400p.
NAKATSUMA, T. Bayesian Analysis of ARMA-GARCH Models: A Markov Chain Sampling Approach. Journal of Econometrics, v.95, n.1, p.57-69, 2000. https://doi.org/10.1016/S0304-4076(99)00029-9
OLIVEIRA, S. C.; ANDRADE, M. G. Comparison between the complete Bayesian method and empirical Bayesian method for ARCH models using Brazilian financial time series. Pesquisa Operacional, v.32, n.2, p.293-313, 2012. https://doi.org/10.1590/S0101-74382012005000019
OLIVEIRA, S. C.; ANDRADE, M. G. Stochastic models with heteroskedasticity: A Bayesian approach for Ibovespa returns. Acta Scientiarum. Technology, v.35, n.2, p.339-347, 2013. https://doi.org/10.4025/actascitechnol.v35i2.13547
SILVA, W.; SÁFADI, T.; CASTRO JUNIOR, L. Uma análise empírica da volatilidade do retorno de commodities agrícolas usando modelos ARCH: os casos do café e da soja. Revista de Economia e Sociologia Rural, v.43, n.1, p.35-42, 2005. https://doi.org/10.1590/S0103-20032005000100007
VEIGA, E. P.; VIVANCO, M. J. F. A medida L como critério de comparação de modelos: Uma revisão da literatura. Revista Brasileira de Biometria, São Paulo, v.30, n.3, p.343-352, 2012.