[Forum SIS] Short course on Bayesian spatio-temporal modelling - Univ. of Perugia

[Maria Giovanna Ranalli]

.: A short course on Bayesian spatio-temporal modelling :.

Instructor 
K. Shuvo Bakar, CSIRO, Canberra

Abstract: 
This course will provide an introduction to basic concepts of Bayesian and 
and spatial/spatio-temporal statistics together with practical implications 
using R software. The course will cover modelling of point reference data
only. However, a brief discussion will be provided for other two types of 
spatial data, e.g., point pattern and areal. No previous knowledge of Bayesian
methods is necessary, however participants should have a good understanding on basic
statistical modelling. 

Place and time:
Dept. of Economics, University of Perugia
Computer Lab of the Statistics Section

Monday, March 10th, 9.00-12.00
Wednesday, March 12th, 9.00-12.00

Short-course outline:
The short course will run for 2-days. Each lecture will be followed up by a practical session 
using R. 

Key topics that will be discussed:
1. Bayesian inference.
   - Theorem, priors, hierarchy, computations.
2. Spatial data and models.
   - Data types, variograms, correlation functions, kriging,
     Gaussian process models, multivariate models. 
3. Spatio-temporal data and models.
   - Gaussian process models, autoregressive models, 
     dynamic models, spatially varying models, 
	 spatio-dynamic models, models for the big-N problem.  
  
Target audience:
Post-graduate researchers including PhD students and early career researchers. 
Researchers from other disciplines with some background in Statistics/Mathematics 
and a strong motivation to learn spatio-temporal methods are also welcome.  

Pre-requisites:
Should have some experience in statistical modelling and some familiarity with R. 
Computers will be available in the lab, but participants can bring their own laptop 
with certain R packages, e.g., spTimer, spBayes, geoR, pre-installed.

References:
(1) Banerjee, S, Carlin, B.P. and Gelfand, A.E. (2004). Hierarchical modelling and analysis for spatial data, Chapman and Hall/CRC.
(2) Cressie, N. and Wikle, C.K. (2011). Statistics for spatio-temporal data. Wiley.
(3) Bakar, K.S. and Sahu, S.K. (2014). spTimer:  Spatio-temporal Bayesian modelling using R. Journal of Statistical Software. In press.

About the Instructor:
Dr Bakar is currently working at CSIRO’s (Commonwealth Scientific and Industrial Research Organisation) Computational Informatics division under the prestigious Office of the Chief Executive  fellowship for his research on environmental and ecological analysis. He has been awarded his PhD by the University of Southampton, UK in 2012. Dr Bakar’s PhD was on developing Bayesian spatio-temporal models for analysing ground level ozone concentrations. As a part of his PhD, he has developed a software package spTimer in R. Dr Bakar’s current work at CSIRO involves developing novel spatio-temporal methods for analysing climatic variables, including data obtained from remote sensing, global climate models etc. He is involved with two research wings at CSIRO, the Climate Adaptation and the Bio-security Flagships. 

Please contact me for any further information,
with best regards

M. Giovanna Ranalli, PhD

~ Associate Professor of Statistics
~ Department of Political Science
~ University of Perugia (Italy) 

~ Tel +39 075 5855245
~ Fax +39 075 5855950
~ url: www.stat.unipg.it/~giovanna
~ Office hours: Mon 12-1 pm; Tue 2:30-3:30 pm


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.stat.unipg.it/pipermail/sis/attachments/20140225/0c626738/attachment-0001.html>