Course of
Data Analysis and Exploration 
Diary of the course 2010/11

  • 17/9. Introduction to the course. Presentation of some statistical ideas (possibly, some notes available shortly).
  • 20/9. Introduction to the use of R R code used.
  • 23/9. Matrices and data frames. First graphical commands R code used.
  • 28/9. Using datasets already within R. Some more graphical commands R code used. Linear models and maximum likelihood estimation.
  • 30/9. Orthogonal projection. Application to the estimation in linear models. variance of the estimator; Gauss-Markov theorem: see slides shown in class (with some improvements). Introduction to linear models in R R code used.
  • 5/10. Multiple regression in R.
  • 7/10. Confidence intervals. Quick review of theory and some examples; computing confidence intervals in R.
  • 12/10. Tests of hypotheses. Quick review of general ideas and theory. Example of the t-test. First tests of hypothesis in R.
  • 14/10. Tests of hypotheses in the linear model. Interpretation of the output of lm() in R. How to perform other tests within the linear model.
  • 26/10. lm() when predictor variables are qualitative. 1-way and 2-way analysis of variance.
  • 28/10. Resolution of some problems in class. Multivariate normal distribution.
  • 2/11. Theory of F tests in linear models. 2-way analysis of variance in R.
  • 4/11. Analysis of covariance. Polynomial regression.
  • 9/11. Regression diagnostics.
  • 11/11. Model selection and cross-validation. See also the slides shown in class
  • 16/11. Simulation of linear models.
  • 18/11. An introduction to generalized linear models and logistic regression. An example of logistic regression in R. Notes by German Rodriguez (thanks) on logistic regression.
  • 23/11. Other examples of logistic and Poisson regression in R.
  • 25/11. Conclusion of examples on Poisson regression in R. Analysis of fit of different models.
  • 30/11. Logistic regression applied to simulated data. Theory of principal component analysis (short summary of principal component analysis, for the moment in Italian) .
  • 2/12. Examples of principal component analysis.
  • 9/12. Linear discriminant analysis.
  • 14/12. Linear discriminant analysis; other examples.
  • 16/12. Discussion of problems, mainly on examples of anova.