Jim Griffin’s research page

I am a Lecturer in Statistics at the School of Mathematics, Statistics and Actuarial Science, University of Kent and an Associate Member of CRiSM, University of Warwick. My contact details are here and my interests are:

·        Bayesian nonparametric methods including posterior simulation and the construction of dependent nonparametric priors.

·        Inference with financial data including the analysis of high frequency data, volatility processes with jumps and the application of Bayesian nonparametric methods to stochastic volatility modelling.

·        Bayesian methods for variable selection methods with many regressors including efficient computation, the use of absolutely continuous priors and applications in bioinformatics.

·        Efficiency measurement using stochastic frontier models.

I’m one of the organisors of a conference on Model Uncertainty which will take place 30th May-1st June, 2010 at the University of Warwick (just before this year’s Valencia conference). Further details, including a stellar list of invited speakers, are available here

 

Preprints                Publications                PhD Students

      


Preprints

Distributions
Modelling overdispersion with the Normalized Tempered Stable distribution (2010, with M. Kolossiatis and M. F. J. Steel)

Markov chain Monte Carlo
On Adaptive Metropolis-Hastings Methods (2009, with S. G. Walker)

Nonparametrics
Posterior Simulation of Normalized Random Measure Mixtures (revised, 2009, with S. G. Walker), Journal of Computational and Graphical Statistics (forthcoming) (Matlab code is available)
Time-Dependent Stick-Breaking Processes (2009, with M. F. J. Steel)
The Ornstein-Uhlenbeck Dirichlet Process and other time-varying processes for Bayesian nonparametric inference (2008)

Regression and Variable Selection
Inference with Normal-Gamma prior distributions in regression problems (revised, 2009, with P. J. Brown), Bayesian Analysis (forthcoming)
Cross-validation prior choice in Bayesian probit regression with many covariates (2010, with D. Lamnisos and M. F. J. Steel)
Bayesian adaptive lassos with non-convex penalization (2007, with P. J. Brown)

Stochastic Frontier Analysis
Bayesian Clustering of Distributions in Stochastic Frontier Analysis (2008)

Volatility Estimation
Covariance measurement in the presence of non-synchronous trading and market microstructure noise (revised, 2009, with R. C. A. Oomen), Journal of Econometrics (forthcoming)


Publications

Nonparametrics
Bayesian Nonparametric Modelling with the Dirichlet Process Regression Smoother (2010, with M. F. J. Steel), Statistica Sinica (forthcoming)
Slice Sampling Mixture Models (2010, with M. Kalli and S. G. Walker), Statistics and Computing, Online First
Default priors for density estimation with mixture models (2010) (Matlab code is available), Bayesian Analysis (forthcoming)
Order-Based Dependent Dirichlet Processes (2006, with M. F. J. Steel), Journal of the American Statistical Association, Theory and Methods, 101, 179-194

Nonparametric Regression
A Bayesian Partition Model for Customer Attrition (2001, with C. J. Hoggart), in Bayesian Methods with Applications to Science, Policy and Official Statistics (Selected Papers from ISBA 2000): The Sixth World Meeting of the International Society for Bayesian Analysis, 223-232

Regression and Variable Selection
Transdimensional sampling algorithms for Bayesian variable selection in classification problems with many more variables than observations (2009, with D. Lamnisos and M. F. J. Steel), Journal of Computational and Graphical Statistics, 18, 592-612 (Matlab Code and a ReadMe file are available)

Stochastic Frontier Analysis
Flexible Mixture Modelling of Stochastic Frontiers (2008, with M. F. J. Steel), Journal of Productivity Analysis, 29, 33-50.
Bayesian Stochastic Frontier Analysis Using WinBUGS (2007, with M. F. J. Steel), Journal of Productivity Analysis, 27, 163-176 (WinBugs code is available)
Semiparametric Bayesian Inference for Stochastic Frontier Models (2004, with M. F. J. Steel), Journal of Econometrics, 123, 121-152 (Matlab code is available)

Volatility Estimation
Bayesian inference with stochastic volatility models using continuous superpositions of non-Gaussian Ornstein-Uhlenbeck processes (2010, with M. F. J. Steel), Computational Statistics and Data Analysis, Article In Press (Matlab code is available)
Sampling Returns for Realized Variance Calculations: Tick Time or Transaction Time? (2008, with R. C. A. Oomen), Econometric Reviews, 27, 230-253.
Inference with non-Gaussian Ornstein-Uhlenbeck processes for stochastic volatility (2006, with M. F .J. Steel), Journal of Econometrics, 134, 605-644

.
PhD Students

Current
Eleni-Ioanna Delatola (University of Kent) – Nonparametric modelling of financial data
Vasiliki Dimitrakopoulou (University of Kent) – Clustering and Variable Selection (with Phil Brown)
Kitty Wan (University of Kent) – Modelling SNP data (with Phil Brown)

Former
Demetris Lamnisos (University of Warwick) – “Bayesian Variable Selection for Binary Regression with Many More Variables than Observations” (with Mark Steel)
Michalis Kolossiatis (University of Warwick) – “Modelling via Normalisation for Parametric and Nonparametric Inference” (with Mark Steel)