Michael K. Pitt

Learn More
In this paper we provide methods for estimating non-Gaussian time series models. These techniques rely on Markov chain Monte Carlo to carry out simulation smoothing and Bayesian posterior analysis of parameters, and on importance sampling to estimate the likelihood function for classical inference. The time series structure of the models is used to ensure(More)
We model a time series fyt; t = 1; :::; ng using a state space framework with the fytj tg being independent and with the state f tg assumed to be Markovian. The task will be to use simulation to estimate f( tjFt), t = 1; :::; n, where Ft is contemporaneously available information. We assume a known `measurement' density f(ytj t) and the ability to simulate(More)
In this paper we develop likelihood based inferential methods for a novel class of (potentially non-stationary) diffusion driven state space models. Examples of models in this class are continuous time stochastic volatility models and counting process models. Although our methods are sampling based, making use of Markov chain Monte Carlo methods to sample(More)
In this paper we provide a unified methodology in order to conduct likelihood-based inference on the unknown parameters of a general class of discrete-time stochastic volatility models, characterized by both a leverage effect and jumps in returns. Given the nonlinear/non-Gaussian state-space form, approximating the likelihood for the parameters is conducted(More)
This paper provides methods for carrying out likelihood based inference on non-linear observed and partially observed non-linear diffusions. The diffusions can potentially be non-stationary. The methods are based on innovative Markov chain Monte Carlo methods combined with an augmentation strategy. We study the performance of the methods as the degree of(More)
In this paper, we provide a method for modelling stationary time series. We allow the family of marginal densities for the observations to be speci ed. Our approach is to construct the model with a speci ed marginal family and build the dependence structure around it. We show that the resulting time series is linear with a simple autocorrelation structure.(More)
This article develops a general-purpose adaptive sampler for sampling from a high-dimensional and/or multimodal target. The adaptive sampler is based on reversible proposal densities each ofwhich has amixture ofmultivariate t densities as its invariant density. The reversible proposals are a combination of independent and correlated components that allow(More)