21210210 - Statistica per big data

The course introduces students to modern data visualization techniques, statistical methods and models for analysing simple and complex dependency structures, and statistical methods for decision support. Some multivariate techniques will also be introduced. The course focuses particularly on exploratory data analysis, linear models and generalized linear models, their strengths and limitations.
Making extensive use of real data examples and their analysis with R and Minitab software, the course will emphasize the role statistical models in addressing scientific questions and how these are translated into relevant statistical questions. The student will learn to distinguish between parameter estimation problems, hypothesis testing and prediction. Therefore, the student will be taught not only to apply statistical techniques but also to choose the most appropriate technique and to comment on the output for decision-making purposes.