A1477
Title: Unified mixture sampler for state-space models: Application to stochastic conditional duration models
Authors: Daichi Hiraki - University of Tokyo (Japan) [presenting]
Yasuhiro Omori - University of Tokyo (Japan)
Abstract: A unified mixture sampler (UMS) is proposed that provides a universal estimation framework for nonlinear state-space models with exponential-exponential likelihood kernels. Unlike existing methods that require deriving new mixture approximations for each specific distribution, the approach dynamically adapts an existing ten-component mixture through a deterministic recentering and rescaling algorithm. Applied to the stochastic conditional duration (SCD) model, the proposed sampler efficiently handles unknown shape parameters such as those in Weibull or Gamma distributions by updating mixture components near-instantaneously during MCMC iterations. The UMS not only simplifies implementation but also ensures exact inference via a lightweight Metropolis-Hastings step. Numerical examples demonstrate that the method substantially outperforms conventional slice sampling approaches, significantly reducing autocorrelation in MCMC samples while maintaining high computational efficiency. This unified framework encompasses a wide range of applications, including logit, Poisson, and various SCD model specifications, providing a highly efficient alternative to model-specific samplers.
