The course aims to introduce students to statistical data analysis. It is intended to provide basic knowledge of inferential statistics.
Expected learning outcomes
At the end of this class, the students are expected to: - Be familiar with techniques and tools for the synthetic and graphical description of the information provided by data sets - Be familiar the language and the models for the representation and the analysis of random phenomena - Know the basics on methods and tools of statistical inference - apply the methods and techniques of statistical analysis to real data sets by means of the use of appropriate statistical software.
Summarizing data and descriptive analysis: types of data, frequency distributions, position and shape indexes (mean, median, quantiles, variance, standard deviation, range, interquartile range,..), histograms, boxplots and other frequency graphs. · Probability and random variables: properties of probability, discrete random variables (Bernoulli, Binomial, Poisson distributions), continuous random variables (Exponential, Gamma, Gaussian, t-Student distributions), mean and variance, properties of means and variances, independence, Central Limit Theorem. · Estimation: sampling distributions, properties of estimators, confidence intervals, testing a hypothesis, principles of significance tests, significance levels and types of error, power of a test. · Comparing samples: comparing the means of two independent Gaussian samples (t-test), comparing the means of two dependent Gaussian samples (paired t-test), comparing two variances of two independent Gaussian samples (F-test), comparing two proportions, comparing several means of independent Gaussian samples using analysis of variance (ANOVA), multiple testing correction. · Regression models: univariate and multivariate linear models, least squares estimators of the parameters of a linear model, tests and confidence intervals for the parameters of a linear model, prediction of a new observation, goodness of fit methods, analysis of the residuals, logistic regression.
Prerequisites for admission
No prerequisites different from those required for admission to the Master Degree program.
Class lectures and practices; during course practices, also given in an informatics room using the student's laptop, the R program language will be illustrated and used.
All the bibliographical suggestions as well as additional material will be available on the "Be e-Poli" (BeeP), the portal for the network activities of students and professors at the Politecnico di Milano, accessible from the Politecnico di Milano Web site; students registered to the course for the current academic year can access it.
Assessment methods and Criteria
The assessment is based on a written exam at the end of the course, with exercises on all the topics presented during the course lectures or practices, and on a data analysis team project evaluation.