On the limited memory BFGS method for large scale optimization
Mathematical Programming: Series A and B
Choice of Basis for Laplace Approximation
Machine Learning
Evaluating derivatives: principles and techniques of algorithmic differentiation
Evaluating derivatives: principles and techniques of algorithmic differentiation
Automatic approximation of the marginal likelihood in non-Gaussian hierarchical models
Computational Statistics & Data Analysis
Computational Statistics & Data Analysis
Hi-index | 0.03 |
A methodology for fitting general stochastic volatility (SV) models that are naturally cast in terms of a positive volatility process is developed. Two well known methods for evaluating the likelihood function, sequential importance sampling and Laplace importance sampling, are combined. The statistical properties of the resulting estimator are investigated by simulation for an ensemble of SV models. It is found that the performance is good compared to the efficient importance sampling (EIS) algorithm. Finally, the computational framework, building on automatic differentiation (AD), is outlined. The use of AD makes it easy to implement other SV models with non-Gaussian latent volatility processes.