[Date Prev][Date Next]
[Chronological]
[Thread]
[Top]
lezioni Prof. Peter Green
Come precedentemente annunciato, si comunica che le lezioni del Prof.
= Peter Green (University of Bristol) si terra' nell'aula 202
dell'Universita' = Bocconi, dalle ore 10.30 alle 12.30 e dalle 14.30
alle 16.30 di domani, mercoledi' 12 maggio.
si allega programma delle lezioni
*******************************
Peter Green, University of Bristol:=20
MCMC methods for nonparametric Bayesian analysis.=20
=20
The unifying theme for these lectures will be the use of MCMC methods
= that handle variable-dimension state spaces to compute posterior =
distributions for functions such as probability densities, point
process = intensities, regression functions and hazard functions.=20
In all of these settings, the use of functional specifications with a
= variable (or random) number of terms provides additional
flexibility, = enhancing the freedom of the data to choose the
functional form that = fits it best. Reversible jump MCMC is the main
tool here, and the = lectures will include an introduction to this
approach to generalising = the Metropolis-Hastings method to general
state spaces.=20
The main statistical topics addressed will include (i) change-point =
analysis for point processes, (ii) analysis of finite mixtures with an
= unknown number of components, (iii) connections between classical =
mixture specifications and Dirichlet process based models, (iv) use of
= mixture models in the Bayesian analysis of factorial experiments,
(v) = applications to inference about hazard functions, and (vi)
applications = to ion channel data.=20
References Arjas, E. and Heikkinen, J. (1997) An algorithm for =
nonparametric Bayesian estimation of a Poisson intensity.
Computational = Statistics, 12, 385--402.=20
Green, P. J. (1995) Reversible jump Markov chain Monte Carlo
computation = and Bayesian model determination, Biometrika, 82,
711--732=20
Green, P. J. and Richardson, S. (1999) Modelling heterogeneity with
and = without the Dirichlet process, available at =
http://www.stats.bris.ac.uk/~peter/Research.html.=20
Hodgson, M. (1999) A Bayesian restoration of an ion channel signal. =
Journal of the Royal Statistical Society Series B, 61, 95--114 Nobile,
= A. and Green, P. J. Bayesian analysis of factorial experiments by =
mixture modelling. available at =
http://www.stats.bris.ac.uk/~peter/Research.html.=20
Richardson, S. and Green, P. J. (1997) On Bayesian analysis of
mixtures = with an unknown number of components (with discussion)
Journal of the = Royal Statistical Society Series B, 59, 731--792=20
=20