Construction of stationary time series via the Gibbs sampler with application to volatility models
Pitt, Michael K. and Walker, S. G. (Stephen G.) (2001) Construction of stationary time series via the Gibbs sampler with application to volatility models. Working Paper. Coventry: University of Warwick, Department of Economics. Warwick economic research papers (No.595).
WRAP_Pitt_twerp595.pdf - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader
Official URL: http://www2.warwick.ac.uk/fac/soc/economics/resear...
In this paper, we provide a method for modelling stationary time series. We allow the family of marginal densities for the observations to be specified. Our approach is to construct the model with a specified marginal family and build the dependence structure around it. We show that the resulting time series is linear with a simple autocorrelation structure. In particular, we present an original application of the Gibbs sampler. We illustrate our approach by fitting a model to time series count data with a marginal Poisson-gamma density.
|Item Type:||Working or Discussion Paper (Working Paper)|
|Subjects:||H Social Sciences > HB Economic Theory
Q Science > QA Mathematics
|Divisions:||Faculty of Social Sciences > Economics|
|Library of Congress Subject Headings (LCSH):||Markov processes, Monte Carlo method, Probabilities, Stochastic processes|
|Series Name:||Warwick economic research papers|
|Publisher:||University of Warwick, Department of Economics|
|Place of Publication:||Coventry|
|Number of Pages:||31|
|Status:||Not Peer Reviewed|
|Access rights to Published version:||Open Access|
Barndorff-Nielsen, O. E. (1997). Normal inverse Gaussian distributions and stochastic
Actions (login required)