Advances and Applications in Statistics
Volume 53, Issue 4, Pages 345 - 361
(October 2018) http://dx.doi.org/10.17654/AS053040345 |
|
BAYESIAN ESTIMATION OF GARCH(p, q) MODEL
Bryan P. Sumalinab and Arnulfo P. Supe
|
Abstract: This paper derive the joint distribution of the parameters of the generalized autoregressive conditional heteroscedastic (GARCH(p, q)) model with student-t innovation using the Bayesian approach. MCMC methods, particularly the Metropolis-Hastings method, were used to find the estimators of the parameters of the model. The efficiency of the estimates is evaluated by simulation study. Further, the performance of the Bayesian estimator was compared to the classical maximum likelihood estimator in terms of mean squared error (MSE). Results show that the Bayesian estimator is more efficient than the classical maximum likelihood estimator. |
Keywords and phrases: GARCH, Bayesian, MCMC, M-H algorithm.
|
|
Number of Downloads: 399 | Number of Views: 3730 |
|