B027530 - STATISTICS FOR SPATIAL DATA

English

The environmental phenomena, but not only, have very often a spatial component that can not be neglected in their analysis.
The aim of the course is to introduce students to techniques statistics collection, description and analysis of spatial data (data where there is a spatial dependence) and to allow students to acquire the technical skills to address the problem of management and development of geographic information.

- Bailey TC, Gatrell AC (1995) Interactive Spatial Data Analysis, Longman.
- Bivand RS, Pebesma EJ, Gomez-Rubio V (2008) Applied Spatial Data Analysis with R, Springer.

The course aims to introduce students to the main techniques used for the analysis of data where it is present to a significant spatial dependence. Moreover, the Geographic Information System is presented.

Preparatory teaching: none formally but Statistical Inference and Statistical Models are suggested.

Lessons of frontal teaching in the classroom and in the laboratory.

e-learning

The assessment consists of an oral test. This test is aimed at verifying the knowledge acquired regarding the concepts, models and tools that have been the subject of the course. Moreover, the ability to apply the acquired knowledge, ability to draw conclusions, communication skills and use of appropriate language, and ability to understand and learning are verified. For the Geographic Information System part a project is evaluated.

1- Introduction to spatial statistics.
2- Stochastic spatial processes and their properties.
3- Point Process data: kernel estimate of the intensity, first-order nearest neighbor distance methods (functions F and G), and K function for the estimation of the intensity of the second order. Marked point processes.
4- Data or geodata random surface: methods for estimate of the area such as moving averages spatial kernel, tessellation. Variogram-covariogram and correlogram models for variogram and covariogram, trend surface analysis, kriging.
5- Area Data: Moran's I for spatial autocorrelation, conditional autoregressive models simultaneous SAR and CAR, CAR Bayesian models. Ecological regression and Geographical Weighted Regression. Small area estimation.
6- Spatial interaction Data: gravity models.
In the laboratory will take presented some libraries of the R software for the description, representation, spatial data analysis.

Introduction to GIS (Geographic Information System) and to the QGIS software.

The 6 CFU exam excludes argument number 3: Point process data.