STATISTICAL LEARNING

Single discipline educational activity

Course Sheet

Course code:

ECBA014

Academic Year of enrolment:

2022

Year of supply:

2023/2024

Type of Course:

Characterising

Degree:

Course of 2nd Cycle Degree in ECONOMICS AND BUSINESS ANALYTICS

Disciplinary Sector:

Statistics

Language:

Italian

Location:

PESCARA

Credits:

Programme Year:

Professor and Collaborators:

IPPOLITI Luigi

Cycle:

First semester

Hours of classroom activity:

Individual study hours:

102

Prerequisites:

Mathematics (calculus), linear algebra, matrices, and statistical inference (estimation and statistical test)

Objectives

The course provides knowledge about the analysis of compex data. The course aims to provide the student with the tools to
extract relevant information from large amounts of data, with particular attention to statistical learning (statistical learning) both in a predictive
and non-supervised context (supervised and non-supervised learning).

LEARNING OUTCOMES
The course aims at completing student's training with notions and tools useful to deepen the aspects of multivariate statistical analysis. The training will then be completed and enriched by the following skills:

Knowledge and understanding / Applying knowledge and understanding
- Knowledge of statistical concepts for multivariate analysis and related specialized terminology
- Ability to apply the principles of statistical reasoning in the preparation and interpretation of company reports
- Ability to use R and MATLAB software for statistical analysis

Making judgements
- To learn the logical and statistical concepts that are indispensable for working independently in the research, selection and processing of
complex data .

Communication skills
- Learn the terminology and statistical techniques of multivariate analysis to communicate or correctly discuss the results of the analysis of
complex data

Contents

The following topics are considered as important parts of the teaching
program for the fulfilment of the objectives: introduction to Statistical Learning, data visualization, regression and classification,
Non-supervised learning (principal component analysis, Clustering, procruste), Analysis if complex data (spatial data and social data mining)

Extended Syllabus

The course aims to introduce methods and models to extract relevant information from large amounts of data, with particular attention to
statistical learning (statistical learning) both in a predictive and nonpredictive context (supervised and non-supervised learning). In order to
provide the skills for the analysis and modeling of real data, the lessons will be supplemented by R exercises in the computer room.

Program:

1. Introduction to data mining e statistical learning.
2. Data Matrix and object oriented data analysis (OODA)
3. Data visualization technique
4. Introduction to probability
5. The multivariate Normal Distribution
6. Prediction model for independent data with R (LDA, K-NN, SVM)
7. Analysis of complex data (Object Oriented Data Analysis) con R
7.1 OODA and Procrustes analysis
7.2 OODA and spatial data analysis
7.3 OODA and Social Data Mining (Text Mining and Natural Language Processing)

Recommended Bibliography

- Slides and handouts for students not attending the course will be available from the professor

- Maindonald, Braun (2010) Data Analysis and Graphics Using R: An Example-Based Approach. 3rd edition, Cambridge University Press
- James, Witten, Hastie, Tibshirani (2013) An Introduction to Statistical Learning (with Applications in R), Springer-Verlag
- Kevin Murphy (2012) Machine learning : a probabilistic perspective, The MIT Press, Cambridge, Massachusetts, London, England
- Luìs Torgo (2011) Data Mining with R. Learning with case studies. CRC Press

Methods of Provision

Conventional

Teaching Methods

Frontal lectures as well as practical exercises with the use of the software R

Evaluation methods

Verification of learning:

The exam is mainly based on an oral discussion aiming at verifying the knowledge of the theoretical part of the topics covered in class. A discussion of a report prepared for the analysis of a data sets (chosen by the student) using the R software is also required. In the determination of the final grade the discussion of the report accounts only for a 30%.

Contacts/More Information

E-mail: ippoliti@unich.it

Students will be received on Mon and Wed between 15:00 and 16:00

Appointments can be fixed by e-mail.

Università degli Studi
"G. d'Annunzio"

Chieti - Pescara

Aree