• Elements of national accounting
• Index numbers
• Productivity and technical efficiency
• Financial ratios and bankruptcy risk prediction
• Statistical methods for forecasting
The first three parts cover basic concepts and methods of economic statistics, while the next two parts address the application of statistical methods and models to business problems.
• O. Castellino. Introduzione alla contabilità nazionale, Giappichelli, 2021.
• A. Predetti. I numeri indici. Teoria e pratica dei confronti temporali e spaziali, Giuffrè, 2006. [Chapters 1, 2, 3, 4.1.1, 4.2]
• T. Laureti. L'efficienza rispetto alla frontiera delle possibilità produttive. Modelli teorici ed analisi empiriche, Firenze University Press, 2006. [Chapters 1, 2, 3]
• L. Biggeri, M. Bini, A. Coli, L. Grassini, M. Maltagliati. Statistica per le decisioni aziendali, Pearson, 2a ed., 2017. [Chapters 5.3, 5.4.1, 8]
• G. James, D. Witten, T. Hastie, R. Tibshirani. Introduzione all'apprendimento statistico con applicazioni in R, Springer, 2a ed., 2009. [Chapters 2.1, 2.2, 3.1-3.4, 4.1-4.3, 5.1, 8.1, 8.2]
• M. Di Nuzzo. Data science e machine learning: dai dati alla conoscenza, 2021. [Chapters 11-14, 19]
• J. C. Hull. Machine learning in business: un'introduzione alla scienza dei dati, 3a ed., 2021. [Chapters 1, 3 (esclusi 3.4-3.8, 3.12), 4, 10]
Learning Objectives
The teaching aims to prepare the student to collect and process useful information for economic analysis and business activity. In order to facilitate understanding of the methods presented, practical applications and real case studies will be proposed in R and Excel. At the end of the teaching, the student will be able to choose and apply appropriate methods in specific contexts, as well as to critically examine the results and to effectively communicate them to specialist and non-specialist interlocutors.
Prerequisites
Knowledge of inferential aspects of linear and logistic regression models.
Teaching Methods
Lectures, illustrations of real case studies in R and Excel, practical exercises.
Further information
Slides and additional materials will be made available on the Moodle platform.
Type of Assessment
Written test + Oral examination
The written test consists of open-ended questions and exercises to be carried out using the scientific calculator. The oral examination consists of theoretical questions on the topics covered in class. Only students who have obtained a sufficient grade at the written test are admitted to the oral examination.
Course program
• 1. ELEMENTS OF NATIONAL ACCOUNTING: introduction to economic statistics; the main economic indicators; the European system of accounts: sequence of accounts, intersectoral tables; impact analysis of final demand on production.
• 2. INDEX NUMBERS: elementary index numbers; synthetic index numbers of prices and quantities: Laspeyres, Paasche and Fisher formulas; consumer price index numbers; value index numbers; inflation rates; deflation of goods or services and of economic aggregates; chained values deflation.
• 3. PRODUCTIVITY AND TECHNICAL EFFICIENCY: partial productivity and total factor productivity (TFP); partial productivity index numbers; TFP index numbers: Laspeyres, Paasche and Fisher formulas; production frontier; output-side and input-side technical efficiency; isoquants; output elasticities; returns to scale; linear production frontier; Cobb-Douglas production frontier; deterministic versus stochastic production frontiers; estimation of technical efficiency from a sample of production units; relationship between productivity and technical efficiency.
• 4. FINANCIAL RATIOS AND BANKRUPTCY RISK PREDICTION: reclassified balance sheet and income statement; main financial ratios; benchmarking; predictive analysis of bankruptcy; construction of a predictive model based on logistic regression; validation and revision of the predictive model based on the ROC curve.
• 5. STATISTICAL METHODS FOR FORECASTING: use of the linear regression model for predicting a quantitative response variable; use of the logistic regression model for binary classification; prediction error: train error versus test error, relationship between prediction error and model complexity, balance between bias and variance; methods to estimate the prediction error: hold-out validation, k-fold cross-validation; methods to select the degree of complexity; other predictive methods: trees and random forests; some applications to business problems.
Sustainable Development Goals 2030
The illustrated case studies aim to support innovation in industry and infrastructure (Goal #9), quality of life in cities and communities (Goal #11), sustainability of production systems (Goal #12).