연구홍보포털

논문 초록 (Abstract)

We develop a new Bayesian Markov Chain Monte Carlo algorithm for Euler-discretized Feller square-root stochastic volatility model s and demonstrate the performance of our algorithm through simulations and empirical analyses. Specifically, our algorithm use the Laplace approximation of the posterior density of conditional variance, which is the probability kernelof the generalized inverse gaussian distribution, derived from the joint density of return and co nditional variance so that it can be easily applied to the extended sto chastic volatility models with such as fat-tailed distributions or L évy jump processes. In addition, we conduct the simulation experiment investigating and comparing the size and power of the parametric specification tests checking certain finite-dimension al moment conditions without correction for parameter estimatio n uncertainty with that of the nonparametric Hong and Li (2005)’s omnib us test which is not affected by parameter estimation uncertainty. The parametric and nonparametric tests are based on the probability inte gral transform of theprediction densities of returns obtained using au xiliary particle filter algorithms. Our experiment result shows that t he classical parametric specification test may have no worse size distortion and better power than Hong and Li (2005)’s test.