55330 - ENVIRONMENTAL STATISTICS

Academic Year 2014/2015

  • Moduli: Fedele Pasquale Greco (Modulo 1) Carlo Trivisano (Modulo 2)
  • Teaching Mode: In-person learning (entirely or partially) (Modulo 1); In-person learning (entirely or partially) (Modulo 2)
  • Campus: Bologna
  • Corso: First cycle degree programme (L) in STATISTICAL SCIENCES (cod. 8054)

Learning outcomes

The course will provide the students with the knowledge of the main theoretical foundations of environmental statistics. The aim is to give the basic tools for the quantitative analysis of environmental data.

Course contents

Part 1 (FRANCESCA BRUNO)

Introduction to environmental Statistics. Introduction to R, a language and environment for statistical computing and graphics. Geostatistics: exploratory analysis of spatial and spatio-temporal data; spatial stochastic processes; moments of a spatial stochastic process, variograms and covariograms; the linear spatial model; parametric forms for the variogram; estimating the variogram; kriging.

 

Part 2 (FEDELE GRECO)

Analysis of areal data:

-Descriptive measures of global spatial association

-Local Indicators of Spatial Association (LISA)

-Statistical tests for global and local correlation

-Areal data models (SAR and CAR)

-Spatial linear regression

Statistical modelling of extreme values: the extremal types theorem; the generalised extreme value distribution; threshold models. 

Readings/Bibliography

  • P.J. Diggle, P.J Ribeiro, Model-based Geostatistics, Springer, 2007.
  • S. Coles An introduction to statistical modeling of estreme values, Springer, 2001.

Teaching methods

Theoretical lectures will be followed by exercises using real data in practical classes.

Assessment methods

The exam is performed jointly over the two parts  of program. It is an oral exam in which the student might present an environmental application developed by using software R. Main attention will be posed on the capability of the student on performing a personal analysis on the dataset assigned. Many attention will be on the statistical language, the presentation of the results and the capability in interpreting the results with respect to the theoretical arguments.

Office hours

See the website of Fedele Pasquale Greco

See the website of Carlo Trivisano