Program
of the course
Data Analysis and Exploration
Academic Year 2010/2011
Multiple linear regression; linear and generalized linear models
Methods of multivariate analysis: principal component analysis, discriminant
analysis, correspondence analysis.
Elements of data exploration and visualization. General strategies and
graphical procedures for data analysis.
Use of specific software. The language R will be used.
References
R.A. Johnson and D.W. Wichern, Applied multivariate statistical analysis, Prentice Hall, 1998 covers completely (with rigour and precision) all
that is discussed in the course
Julian J. Faraway, Practical Regression and Anova using R. available at http://cran.r-project.org/doc/contrib/Faraway-PRA.pdf
is a simpler reference for
the first part of the course, that includes detailed instructions on how to
perform the analysis in R.
Heiberger, R.M. and Holland,B. Statistical analysis and data display, Springer, New York, 2004 discusses in a practical way many statistical
problems, with examples of the use of different statistical software including
R. It also recalls quickly the basic concepts and methods of statistics.
Flury B. A first course in multivariate statistics. Springer, New York, 1998 is an alternative reference for the second part
of the course, although it does not cover all the material.
Venables W.N, Ripley B.D. Modern Applied Statistics with S-PLUS. Springer, New York, 1997 is a
bible on how to do a lot of complicated analysis using S-plus or R ; it does
not provide many explanations, though.
F. Crivellari, Analisi statistica dei dati con R, Apogeo, Milano, 2006 is a
simple text (in Italian) that teaches R from the beginning, and uses it to
perform some simple statistical analysis.
Additional references
W. Hardle, L. Simar, Applied multivariate Statistical Analysis, Springer 2012 is a
rigorous textbook that covers all topics of multivariate analysis seen in the course, and also, briefly, statistical ideas and regression models.
B. Everitt, T. Holborn, An Introduction to Applied multivariate Analysis with R, Springer 2011 is a
simple textbook that gives the basic ideas of most topics in multivariate analysis, and also show R code to use in some examples.
Both these books are available (up to January 31, 2013) as E-books from the computers on the University network, thanks to a trial agreement with Springer (for details)