[Date Prev][Date Next] [Chronological] [Thread] [Top]

No Subject

------------------------------------------------------------------------------
                      UNIVERSITA' L. BOCCONI

                ISTITUTO DI METODI QUANTITIATIVI

------------------------------------------------------------------------------

Nell'ambito del ciclo di lezioni su:

         METODI NON PARAMETRICI PER L'ANALISI DELLA SOPRAVVIVENZA


mercoledi' 14 aprile 1999, ore 10.00, in aula 32, (anziche' aula 4 come
inizialmente previsto) si  terra' la  terza lezione dal titolo


   INTRODUCTION TO MCMC METHODS, WITH APPLICATION TO SURVIVAL ANALYSIS
                      
                         Stephen Brooks
                      University of Bristol


Markov Chain Monte Carlo (MCMC), as a computer-intensive statistical tool,
has enjoyed an enormous upsurge in interest over the last few years. The
aim of these lectures is to provide a simple, comprehensive and tutorial
review of the ideas and application of these methods with regard to
survival analysis.

After a brief revision of Markov chains and the Monte Carlo method, we
begin by discussing the two most common MCMC algorithms, namely the Gibbs
sampler and the Metropolis Hastings algorithm, before progressing to newer
and more complex algorithms, such as the slice sampler and
dimension-jumping methods.
We also discuss the implementation of these algorithms. We take a look at
the arguments for and against multiple replications, consider how long
chains should be run for, and how to determine suitable starting points. We
also take a look at graphical models and how graphical approaches can be
used to simplify MCMC implementation.

Finally, we present a couple of examples, which we use as case studies to
highlight some of the points made earlier in the lectures.  We use these
examples both to illustrate the  methodology but also to illustrate the use
of BUGS a simple-to-use package for implementing MCMC.
                  


PER ULTERIORI INFORMAZIONI e  RIFERIMENTI BIBLIOGRAFICI,  
CONSULTARE LA  PAGINA WEB 

http://www.uni-bocconi.it/corsoimq.htm 



Per chiarimenti aggiuntivi e' possibile scrivere  a 

seminari.imq@uni-bocconi.it



 	

 



Pietro Muliere
   Dip. di Economia Poltica e Metodi Quantitativi
   Via San Felice 5, 27100 PAVIA - ITALIA
   Tel +39 - 382 - 506 213
   Fax +39 - 382 - 304 226