Analytics and Data Science in Economics and Management I
Python Course for Data Science (M. Puliga):
- Introduction to the language: basic statements, cycles and functions
- Diving into the language: advanced types: sets and dictionaries, classes and modules, using PIP and ipython
- Scraping the web: introduction to BeautifulSoup, the regular expressions module re, the request module
- Introduction to Plotting: basic numpy, plotting overview
- Data science utilities: introduction to SQL (sqlite/mysql)

Getting, Organizing and Analyzing the Data (A. Petersen):
- a crashcourse in scraping and parsing XML based websites (in mathematica & python)
- a crashcourse on using Gephi (open source network visualization program)
- a crashcourse on data organization, processing, analysis and visualization

Multivariate Statistical Analysis (M. Di Lascio):
This part of the course aims at introducing the most important and widespread multivariate analysis methods. After introducing multivariate random variables and multivariate data, the course focuses on the study of the relationships between variables and on the study of the similarities between units. In particular, the course deals with:

- dimension reduction methods (principal component analysis and factor analysis);
- analysis tools for linear and nonlinear multivariate dependence (canonical correlation analysis and copula models); distance?based and model?based clustering techniques.

Causal Inference and Evaluation Methods (F. Mealli):
This part of the course deals with statistical methods for inferring causal effects from data from randomized experiments or observational studies. Students will develop expertise to assess the credibility of causal claims and the ability to apply the relevant statistical methods for causal analyses.

Prerequisites: Lin. Alg. + Opt. Control + Found. of Prob. & Stat. Inf.
Alexander Petersen (IMT Lucca), Fabio Pammolli (Politecnico di Milano), Marta Di Lascio (Libera Università di Bolzano), Michelangelo Puliga (IMT Lucca), Fabrizia Mealli (Università degli Studi di Firenze)
Analytics and Data Science in Economics and Management II
Empirical studies on heterogeneous firms (F. Pammolli, A. Rungi):
This part of the course aims at providing students with hands-on empirical tools to test the behaviour of economic agents that are heterogeneous in nature. How productive is a firm, an industry or a country? Why? Where is it more profitable to locate an economic activity? How long can we expect a company to outlive its competitors? After introductions to up-to-date illustrative contributions to economic literature, students will be asked to perform their own analyses and comment results after applications to micro data provided during the course. The objective is to develop a critical understanding of the iterative research process leading from real economic data to the choice of the best tools available from the analyst kit.

Business analytics (M. Riccaboni):
The goal is to teach the students how to produce an empirical paper in business research using quantitative data. We cover some of the most widely used methodologies. The course will bridge the gap between applications of methods in published papers and practical lessons for producing your own research.

TBD (J. Marcucci)

Prerequisites: Analytics and Data Science in Economics and management I + Econometrics
Massimo Riccaboni (IMT Lucca), Fabio Pammolli (Politecnico di Milano), Armando Rungi (IMT Lucca), Juri Marcucci (Banca d&#039, Italia)
Applied Econometrics
This course deals with the following topics:

1) Regression and Causality: a) Properties of the Conditional Expectation Function; b) Bad controls; c) Omitted variable bias; d) Measurement errors; e) Simultaneous equations; f) How to write an empirical project.

2) The Evaluation Problem and Randomised Experiments: a) Introduction to the evaluation problem; b) Randomised Experiments; c) Practical problems when running experiments; d) Duflo et al (2007) on randomization in development; e) Application I: Krueger (1999) on class size and educational test scores; f) Application II: Blundell et al (2004) on education and earnings in the UK.

3) Quasi-Experiments: a) Matching; b) Propensity Score Matching; c) Evaluating the validity of matching estimators; d) Application I: Caliendo et al., (2005) on job creation in Germany; e) Application II: Jones and Olken (2009) on assassination and institutions.

4) Differences-in-Differences: a) Basics; b) Regression Differences-in-Differences; c) The Synthetic Control Method; d) Application I: Card & Krueger (1994) on minimum wage and unemployment; e) Application II: Abadie & Gardeazabal (2003) on the effect of terrorism in the Basque region; f) Application III: Autor (2003) on unjust dismissal doctrine and employment.

5) Regression Discontinuity Design: a) Sharp RD; b) Fuzzy RD; c) Running RD Models; d) Application I: Lee (2008) on U.S. House elections; e) Application II: Angrist & Lavy (1999) on scholastic achievement.

Prerequisites: Econometrics
Vincenzo Bove (University of Warwick)
Banking and Finance (long seminar with optional exam)
One of the most challenging task in finance is the gap between theoretical models and the actual software implementation. Cross some different areas (derivatives evaluation, risk management, accounting issues) several problems arise: discretization, analytical approximation, montecarlo simulation vs. numerical probability, optmization and so on. After a short overview of the main financial areas, the course aims to give some insights on these topics, with a special focus on the risk management current hard problems and the related software algorithms.

Prerequisites: Stochastic Processes and Stochastic Calculus, Management Science and Corporate Finance, Finance
Michele Bonollo (IASON ltd.)
Basic Numerical Linear Algebra
The course is aimed to recall the basic notions about vectors, matrices, vector spaces and norms, along with the basic numerical methods concerning the solution of linear systems. In particular: direct methods for square linear systems and conditioning analysis; direct methods for solving over-determined linear systems in the least square sense. The course also provides an introduction to Matlab, which is used for implementing the illustrated methods.
Luigi Brugnano (Università degli Studi di Firenze)
Convex Optimization
The course covers the basics of convex optimization methods, with an emphasis on numerical algorithms that can solve a large variety of optimization problems arising in control engineering, machine learning, mechanical engineering, statistics, economics, and finance.

The materials for the course are available at
Stephen Boyd (Stanford University), Steven Diamond (Stanford University), Enzo Busseti (Stanford University)
Critical Thinking (long seminar without exam)
Critical Thinking is an introductory course in the principles of good reasoning. Its main focus lies in arguments, their nature, their use and their import. Unlike a course in pure Logic, which would spell out universal formal rules of correct reasoning, Critical Thinking is more concerned with the unruly nature of real argumentation that does not allow unambiguous and definite formalization. The course is designed to serve as a methodical preparation for more effective reasoning and improved cognitive skills. Its ambition is to develop those intellectual dispositions that are essential for effective evaluation of truth claims as well as for making reasonable decisions based on what we know orbelieve to know. It is more about the quality of our beliefs and the reasons that support them than about their content. It will make ample use of examples taken from real world case studies, books, scientific or newspaper articles. Students will be encouraged to participate in the discussion over each example, and to find out more of their own.
Stefano Gattei (Chemical Heritage Foundation, Philadelphia)
Data Analysis and Visualization
This course covers some of the most important methodological issues arising in any field of applied economics when the main scope of the analysis is to estimate causal effects. A variety of methods will be illustrated using theory and papers drawn from the recent applied literature. The aim is to bridge the step from a technical econometrics course to doing applied research. The emphasis will be on the applications. The goal is to provide students with enough knowledge to understand when these techniques are useful and how to implement each method in their empirical research.

Part 1 (C. Tealdi):
- The simple regression model
- Estimation
- Inference
- Dummy variables
- OLS Asymptotics
- Heteroskedasticity
- Instrumental Variables

Part 2 (A. Belmonte):
- A quick introduction to STATA
- Microeconomic data structures
- Conditional distributions and the conditional expected function concept
- A discussion of the assumptions of the GM Theorem
- Heterogeneous conditional distributions
- Dummy variables and Anova models
- Autocorrelation and the Moulton factor

Part 3 (P. Zacchia):
- Beyond Single-Equation Linear Models: Structural Models, Identification and Causality; Rubin Causal Model
- Simultaneous Equation Models
- Introduction to M-Estimation
- Generalized Method of Moments
- Maximum Likelihood Estimation
- Non-Parametric Estimation

Part 4 (A. Rungi):
- Benefits and limits of panel data structures; Review of some panel data examples for micro- and macro data; How to handle and describe panel data; basic estimation techniques
- Details for panel data estimators
- Identification problems
- Introduction to non-linear panel data estimators
- Count data models

Prerequisites: Linear Algebra + Found. of Prob. & Stat. Inf.
Cristina Tealdi (Heriot-Watt University), Armando Rungi (IMT Lucca), Paolo Zacchia (IMT Lucca), Alessandro Belmonte (IMT Lucca)
Foundations of Probability and Statistical Inference
This course aims at introducing, from an advanced point of view, the fundamental concepts of probability and statistical inference.
Some proofs are sketched or omitted in order to have more time for examples, applications and exercises. In particular, the course deals with the following topics:

? probability space, random variable, expectation, variance, cumulative distribution function, discrete and absolutely continuous distributions, random vector, joint and marginal distributions, joint cumulative distribution function, covariance,
? conditional probability, independent events, independent random variables, conditional probability density function, order statistics,
? multivariate Gaussian distribution,
? probability-generating function, Fourier transform/characteristic function,
? types of convergence and some related important results,
? point estimation, interval estimation, hypothesis testing, linear regression, introduction to Bayesian statistics.
Irene Crimaldi (IMT Lucca)
Funding and Management of Research and Intellectual Property (long seminar without exam)
The long seminar aims at providing an overview on the management of intellectual property rights (copyright transfer agreements, open access, patents, etc.). Funding opportunities for PhD students, post-docs, and researchers are also presented (scholarships by the Alexander von Humboldt Foundation; initiatives by the Deutscher Akademischer Austausch Dienst; scholarships offered by the Royal Society in UK; bilateral Italy-France exchange programmes; Fulbright scholarships; Marie Curie actions; grants for researchers provided by the European Research Council). For each funding scheme, specific hints on how to write a proposal are given.
Marco Paggi (IMT Lucca), Tbd
Identification, Analysis and Control of Dynamical Systems
The course provides an introduction to dynamical systems, with emphasis on linear systems. After introducing the basic concepts of stability, controllability and observability, the course covers the main techniques for the synthesis of stabilizing controllers (state-feedback controllers and linear quadratic regulators) and of state estimators (Luenberger observer and Kalman filter). The course also covers data-driven approaches of parametric identification to obtain models of dynamical systems from a set of data, with emphasis on the analysis of the robustness of the estimated models w.r.t. noise on data and on the numerical implementation of the algorithms.
Alberto Bemporad (IMT Lucca), Dario Piga (IMT Lucca)
Introduction to Networks
The course will provide an introduction to the mathematical basis of Complex Networks and to their use to describe, analyze and model a variety of physical and economic situations.


Lecture 1 Graph Theory Introduction:
Basic Definitions, Statistical Distributions, Universality, Fractals, Self-Organised Criticality

Lecture 2 Properties of Complex Networks:
Scale-Invariance of Degree Distribution, Small-World Effect, Clustering

Lecture 3 Applications:
Internet, WWW, Socio-technological systems, Economics, Biology

Lecture 4 Communities:
Community Detections, Algorithms to explore Graphs

Lecture 5 Different kind of graphs:
Vertices differences, Layered Vertices, Trees and Taxonomies

Lecture 6 Ranking:
Hierarchies, Spanning Trees,HITS, PageRank,

Lecture 7 Static Models of Graphs:
Erdos-Renyi, Small World,

Lecture 8 Dynamical Models of Graphs:
Barabasi-Albert, Configuration models

Lecture 9 Fitness models:
Fitness model and Self-Organised Fitness Model

Lecture 10 Basic Ingredients of Models:
Growth Preferential Attachments, Log Normal Distribution, Multiplicative Noise
Guido Caldarelli (IMT Lucca)
Machine Learning and Pattern Recognition
Basics of pattern recognition and machine learning and real world applications in imaging, internet, finance. Similarities and differences. Supervised vs unsupervised learning. Linear regression in many ways. The logistic regression. Support vector machines for classification and regression. Random Forests for classification. Linear and quadratic discriminant analysis. Unsupervised learning (k-means, c-means, kernel k-means, spectral clustering, EM). Feature extraction and selection (PCA, ICA, kernel PCA, and manifold learning). Current trends in Machine Learning.
Prerequisites: Probability and basic random processes, linear algebra, basic computer programming, numerical methods.
Sotirios Tsaftaris (The University of Edinburgh)
The sequence in macroeconomics will introduce students to the literature that studies the aggregate evolution of the economy both in the short and long run. A particular emphasis will be given to the role of institutions in explaining economic performance in the long run. The role of monetary policies for the short-run evolution of the economic cycle will be addressed in the last module of the sequence.

Part 1 (D. Ticchi)
-Traditional Keynesian Theories of Fluctuations
-The Lucas Imperfect-Information Model and The Lucas Critique
-The Solow Model

Part 2 (M. Onorato)
-The Neoclassical Growth Model
-Growth with Overlapping Generations. Social Security and Capital Accumulation
-First-Generation Models of Endogenous Growth
-Growth Empirics
-Fundamental Determinants of Differences in Economic Performance

Part 3 (A. Paccagnini)
-The nature and function of money. Classical monetary theory, neutrality and inflation. RBC Model
-Empirical evidence on money. Identification monetary shocks. VAR and SVAR models
-The Basic New Keynesian Model: main ingredients
-Introducing medium-scale DSGE: the Smets-Wouters model
-Financial Frictions in DSGE
-Estimation of DSGE models: some basic concepts
-Unconventional Monetary Policy and Zero Lower Bound: Forward Looking and Quantitative Easing
Davide Ticchi (Università Politecnica delle Marche), Massimiliano Gaetano Onorato, Alessia Paccagnini (University College Dublin)
Management Science and Corporate Finance
The first part of the course is also designed for the curriculum AMCH as ?Basics of Management? (20 hours).

The course aims at introducing doctoral students to fundamental concepts and to theoretical research in management science and corporate finance. The first classes will focus on a few classic readings in management and organization theory, and then we will move to corporate finance. The main goal is to expose students to classic and fundamental background readings on the structure of Organizations, Decision Making in Organizations, and Corporate Finance. Specific Chapters will be selected for class discussions.

Prerequisites: The course will emphasize intuition over technical detail wherever possible, while more technical readings will be made available and discussed with students with a quantitative background. Those who are willing to catch up with some reading on their own should not have too many problems, under the Lecturer's supervision.
Fabio Pammolli (Politecnico di Milano), Luca Regis (IMT Lucca)
Marketing Science and Consumer Behavior
The main goals of the course are:

(1) to take economic theories and methodologies out into the world, applying them to interesting questions of individual behavior and societal outcomes;
(2) to develop a basic understanding of human psychology and social dynamics as they apply to marketing contexts;
(3) to become familiar with the major theory and research methods for analyzing consumer behavior;
(4) to develop market analytics insight into consumer actions.

Most of time will be devoted to close reading of research papers, including discussion of the relative merits of particular methodologies. Students will participate actively in class discussion, engage with cutting-edge research, evaluate empirical data, and write an analytical paper. The course aims at enabling students to develop and enhance their own skills and interests as applied microeconomists.
Massimo Riccaboni (IMT Lucca)
The course deals with some fundamental topics in Microeconomics. It aims at bringing the students from an intermediate to an advanced level of exposure and understanding of the material. The course will give emphasis to problem solving. For this reason problem sets will be assigned during the course at dates to be communicated in class. Students will then rotate on the board in a following lecture to discuss the problems. Problem sets will not be marked but discussion will be taken into consideration in the final evaluation (30%). A final, written, 3 hours test will finalize the evaluation (70%).
Topics: Consumer theory; Production theory; General equilibrium; Expected utility and choice under uncertainty; Contract theory and asymmetric Information; Mechanism design (if time).
Nicola Dimitri (Università degli Studi di Siena)
The course is structured into three modules: the first one will cover advanced topics in complex network theory, whereas, the second one will focus on economic and financial networks, dealing with both theory and applications.

Module 1: Advanced Theory of Complex Networks
Lecture 1 Models of Evolving Networks
Lecture 2 Fitness & Relevance models
Lecture 3 The Master Equations approach
Lecture 4 Percolation
Lecture 5 Epidemic Models on Networks
Lecture 6 Advanced Topological Properties
Lecture 7 Complex Networks Randomization
Lecture 8 Exponential Random Graphs
Lecture 9 Parameter Estimation via Maximum Likelihood
Lecture 10 Applications: Bipartite, Directed and Weighted Networks.

Module 2: Economic & Financial Networks
Lecture 1 Evolutionary Network Games
Lecture 2 Heterogeneous Mean-Field Theory
Lecture 3 Financial Networks
Lecture 4 Systemic Risk
Lecture 5 DebtRank
Lecture 6 Economic Networks
Lecture 7 The WTW & COMTRADE dataset
Lecture 8 Gravity Models of Trade
Lecture 9 Early Warning Signals
Lecture 10 Network Reconstruction from Partial Information

Module 3 Social and Infrastructural Networks
Lecture 1 Introduction to Social Network Data
Lecture 2 Tecniques and Methodologies of Analysis in Social Networks
Lecture 3 Twitter data and Models
Lecture 4 Clustering and Classification of Facebook Data
Lecture 5 Automatic Topic Extraction
Lecture 6 Introduction to Infrastructural Networks
Lecture 7 Electric Grids
Lecture 8 Cascade Phenomena
Lecture 9 Modelling of infrastructural networks
Lecture 10 Smart Grids and Renewables

Prerequisites: Linear algebra, Introduction to Networks, Found. Prob. & Stat. Inf.
Guido Caldarelli (IMT Lucca), Antonio Scala (CNR - Istituto di Sistemi Complessi), Tiziano Squartini (IMT Lucca), Giulio Cimini (IMT Lucca), Fabio Saracco (IMT Lucca)
Numerical Methods for the Solution of Partial Differential Equations
The course introduces numerical methods for the approximate solution of initial and boundary value problems governed by linear partial differential equations (PDEs) ubiquitous in physics, engineering, and quantitative finance. The fundamentals of the finite difference method and of the finite element method are introduced step-by-step in reference to exemplary model problems related to heat conduction, linear elasticity, and pricing of stock options in finance. Notions on numerical differentiation, numerical integration, interpolation, and time integration schemes are provided. Special attention is given to the implementation of the numerical schemes in Matlab and in the finite element analysis program FEAP fast intensive computations.
Marco Paggi (IMT Lucca)
Optimal Control
Discrete-time optimal control: dynamic programming for finite/infinite horizon and deterministic/stochastic optimization problems. LQ and LQG problems, Riccati equations, Kalman filter. Deterministic continuous-time optimal control: the Hamilton-Jacobi-Bellman equation and the Pontryagin?s principle. Examples of optimal control problems in economics.
Giorgio Stefano Gnecco (IMT Lucca)
Philosophy of Science (long seminar without exam)
We know a lot of things ? or, at least, we think we do. Epistemology is the branch of philosophy that studies knowledge: its main features, the dynamics of its growth, as well as its claims for truth, validity, and progress. In this course ? which is designed as a series of seminars held by the students, preceded by a few introductory lectures ? we will consider some of the key contributions to the philosophical debate about the growth of scientific knowledge in the twentieth century, from Logical Positivism to Karl Popper, from Thomas Kuhn to Paul Feyerabend. We shall read some of their (as well as others?) works, and critically consider the content and limits of the different methodologies they advanced.
Finally, we will reflect on the extent to which such debates affected the methodology of the social sciences, and consider in what ways hard and social sciences differ: as to their inner nature, the context in which they operate, the data they employ and rely upon, and the prescriptive methodology they more or less explicitly adopt.
Stefano Gattei (Chemical Heritage Foundation, Philadelphia)
Project Management
Project management; event management; communication and marketing; practical tools of organization; budgeting. Dealing with multiple stakeholders/ Risk management / Time management / PM tips to run an international research/Management plan concept on heritage sites / When applying for funds how do we measure project success / How we manage the output of the management plan / Flat organisations.
Beatrice Manzoni (SDA Bocconi School of Management)
Scientific Writing, Dissemination and Evaluation (long seminar without exam)
In order to ensure their widest possible dissemination, research results need to be presented in academic publications and in talks. The first goal of this course is to introduce students to basic principles of academic writing and on basic techniques to plan and deliver good academic talks. In addition, the course discusses the key principles of peer review, which is what makes science reliable knowledge. In particular, the course focuses on how to write a professional referee report.

Further information is available at
Luca Aceto (Reykjavik University)