Course Topics
- Time series analysis, modeling and forecasting - Dimensionality reduction techniques: principal component analysis - Identifying underlying structures: factor analysis - Unsupervised learning: distance-based clustering algorithms - Model validation and re-sampling - Applications with the software R and the programing language Python

Propaedeutic courses
Prerequisites It is highly recommended both Mathematics for EPE and Statistics for EPE.

Teaching format
Lectures and laboratory sessions.

Educational objectives
Knowledge and understanding: At the end of the course, students will have acquired the following knowledge and understanding: - knowledge of mathematical techniques for solving optimisation problems; - knowledge of probabilistic and inferential tools that allow the use of statistical models; - ability to model social and economic phenomena; - ability to give an economic interpretation to the results of the various mathematical-statistical models applied to economics; - basic knowledge of data management and computer programming for the statistical and econometric analysis of socio-economic data; - knowledge of the technical vocabulary of the subjects taught in this area of learning.. Applying knowledge and understanding: - Ability to interact fluently and spontaneously with native speakers on economic topics; - ability to independently analyse data and identify and explain relationships between real phenomena: - ability to manage simple databases and perform socio-economic data analysis with the support of software; - ability to use quantitative methods to solve economic problems; - ability to read, write and communicate in the technical language of quantitative methods in the three official languages of instruction. Autonomy of judgement Acquisition of the capacity for judgement and methodological tools useful for the critical analysis of data, sources, assumptions and implications of scientific practice, and of the political, ethical and legal context within which economic phenomena are embedded and with which they interact Communication skills Fluency (oral and written) in Italian, German and English, including translation between these languages. Intercultural competence. Conceptual awareness, ability to summarise and express oneself in writing, particularly with regard to the drafting of scientific or science-based documents Learning skills Promotion of critical thinking and analytical skills to focus on complex problems in their long-term dynamics and in the variety of their implications, including ethical ones

Additional educational objectives and learning outcomes
Knowledge and understanding of multivariate data and time series data and several techniques to analyze them and get information on the phenomena of interest. Applying knowledge and understanding of advanced quantitative methods to describe and analyze economic and social phenomena through statistical software. Making judgments on models and statistical tools useful for advanced data analysis. Communication skills in presenting in a correct and concise way methods and results of a statistical analysis. Learning skills of different statistical methods useful in data science.

Assessment
Written exam with theoretical questions, problem-solving exercises, and interpretation of analysis results in R or Python. A voluntary midterm and obligatory final exam, both written. The midterm grade can be rejected in which case you will take the full final exam. This exam format applies to both attending and non-attending students.

Evaluation criteria
Attending and non-attending students 100% written exam consisting of theoretical questions and data analysis tasks. The final grade will be a weighted average of the written midterm exam (50%) and the written final exam (50%). Students who do not take the midterm or reject their midterm grade will be given a longer exam that will count for 100% of the final grade. Criteria for written exam: correctness and clarity of answers, knowledge and understanding of statistical methods, ability to interpret outputs and to correctly use formal code.