DescriptionIn many scientific and other contexts, there is a natural spatial component to data. For example, we may observe the locations at which cases of a particular disease were observed or we may record at several locations the thickness of a layer of rock containing a valuable mineral or we may have satellite imagery recording variation in the amount of heat/light emitted over a region of the earth, possibly at several wavelengths . The structure of such data varies greatly: in the first of the examples, we simply observe the locations; in the second we may control the locations and observe the thickness; the third a rectangular region is divided into small squares and we record a measure of the total radiation from each square. It is usually important to take the spatial component into account when building statistical models for such data. A great many different modeling approaches have been developed. Examples include point processes, marked point processes, geostatistics and Markov random fields. Each modeling approach tackles the important features of a particular kind of spatial data: each of the three examples needs a different kind of tool. After developing an understanding of the basic concepts and fundamental statistical models for spatial data, students may specialise into particular models and/or applications. It is expected that students will use appropriate software, such as R, to bring the topic to life. Prerequisites Statistical Concepts II It is strongly recommended that students doing this project should also have taken Statistical Methods III. Some possible sub-topics may also benefit from having taken Bayesian Statistics III or by taking Topics in Statistics IV. ResourcesThe following are a few of the many books on various aspects of the topic. Diggle and Ripley are classics and quite approachable. Cressie is a well-respected reference for many aspects of the topic.Cressie N, Statistics for Spatial Data, Wiley 1991 (ISBN 0471843369). There is also a more recent edition which is on order for the library. Diggle PJ, Statistical analysis of spatial point patterns, Arnold 2003 (ISBN 0340740701) Ripley BD, Spatial Statistics, Wiley 1981 (ISBN 0471083674) You might also like to look at a short introduction to some ways of doing spatial analyses using R |
email: P S Craig