Official statistics for socio-economic analysis. Publications, databases and informative systems produced by ISTAT. Simple and multiple regression analysis. Introductory spatial data analysis. Laboratory exercises with R and GeoDa.
Lecture notes and teaching materials are available on the Moodle platform.
Learning Objectives
The first part of the course aims to provide students with a knowledge and understanding of the sources of official statistics for socio-economic analysis. Special attention will be placed on the quality of data.
The second part of the course covers the theory and practice of linear regression models, with a brief introduction on spatial autocorrelation and spatial regression models.
During the lab sessions the students will acquire the knowledge to use R and GeoDa to estimate and evaluate different regression models.
Prerequisites
Basic oncepts in descriptive and inferential statistics.
Teaching Methods
Classroom lectures and computer labs.
Further information
To access the Moodle page students have to send a request to the teacher from their UNIFI mail address.
Type of Assessment
The oral examination shall cover the subjects described in the next section (at least one question for each of the first seven subjects).
Course program
1. STATISTICAL DATA FOR ECONOMIC ANALYSIS. Official statistics. The quality of statistical processes and products. Open data. The Italian National Institute of Statistics (ISTAT) and the National Statistical System (SISTAN). ISTAT products: databases, publications, microdata files. Big data in official statistics. Territorial statistics. NUTS. Other sources of statistical data.
2. SIMPLE LINEAR REGRESSION. Descriptive approach. Least squares regression. Goodness of fit and the coefficient of determination. Inferential approach. Properties of regression coeffcients and hypothesis testing. The Gauss-Markov theorem.
3. MULTIPLE REGRESSION MODEL. Interpretation of the multiple regression coefficients. Partial correlation. Multicollinearity. Dummy variables.
4. MODEL MISSPECIFICATION. Heteroskedasticity. Autocorrelation. Leverage points, outliers, influential points.
5. SPATIAL ANALYSIS. Introductory concepts. Exploratory spatial data analysis. Graphic tools.
6. SPATIAL AUTICORRELATION. Contiguity matrix. Global measures: Moran's I, Moran scatterplot, Geary's C and Getis-Ord statistics. Local measures: LISA.
7. SPATIAL REGRESSION MODELS. Spatial error model. Spatial lag model. Diagnostic tests for spatial dependence.
8. Lab sessions with applications of R and GeoDa on sample data sets.