SIdE Postgraduate Courses
August, 28 - September, 1, 2023, Venice, Italy
August, 28 - September, 1, 2023, Venice, Italy
Coordinator: Gaetano Carmeci, DEAMS, Trieste University. E-mail: gaetano.carmeci@deams.units.it
Instructors: Roberto Casarin (University Ca' Foscari of Venice), Matteo Ciccarelli, European Central Bank, Federico Bassetti (University Polytechnic of Milan)
The summer school is organized in collaboration with the Venice Center for Risk Analytics for Public Policies (VERA) and the Italian Society of Econometrics (SIdE).
The course is an introduction on Bayesian Inference, starting from first principles and covering topics of interest for applied econometricians in economics and finance. The course is addressed to students without previous knowledge of Bayesian Econometrics. The methods introduced in the lectures will be illustrated with hands-on applications in MATLAB and R based on reasoned statistical and economic examples.
Program and Syllabus: Link to file pdf
Timetable: Link to file xlsx
Softwares: Prior to the beginning of the course you should install and familiarize yourself with MATLAB and R.
Zoom: Link to SIdE Zoom Classes. Instructions on how to connect to the webinars via Zoom and rules are available at this link. To participate you don’t need to download or buy Zoom. It will be enough to click on the link and follow the instructions on the screen. Please be reminded that recording lectures is not permitted. Access will be denied to non-registered participants.
Office Hours: Scheduled between 14.00 and 15.00. To attend office hours use a private room "Office Hour Room" in the main Zoom session.
BEAR Toolbox (A comprehensive VAR toolbox. )
Econometrics Links Page (Resources and libraries for different softwares)
Chris Sims's Home Page (Codes and libraries on Bayesian inference from his book)
Gary Koop's Home Page (Matlab codes and references on Stochastic Search, Bayesian VAR,...)
Dimitris Korobilis' Home Page (Matlab and references)
Fabio Canova's Home Page (Matlab and empirical macro)
Joshua Chan's Home Page (Matlab and R codes and further references)
LeSage's Home Page (Matlab codes and libraries on Bayesian inference): Spatial Econometrics
Tobias's Home Page (Matlab codes and libraries on Bayesian inference from his book)
Mark Glickman's Page on Bayesian Music
1 Bayesian Inference
1.1 Fundamentals of Bayesian Statistics (link to pdf, imgInvGa, imgGa)
Introduction to Bayesian inference and foundation of decision theory.
Assignment 1: (Text, Solution)
1.2 Review of Bayesian estimation (link to pdf)
Introduction to the Bayesian linear regression model with natural conjugate prior.
Assignment 2: (Text, Solution)
2 Bayesian computation
2.1 Introduction to Monte Carlo methods (link to pdf/code)
Posterior approximation, application to linear and nonlinear models
2.2 Filtering methods (link to pdf/code)
Introduction to filtering methods, univariate and multivariate SV models
Assignment 3: (Text, Solution)
3 Bayesian methods for regression models
3.1 Introduction (link to pdfA/codeA, link to pdfB/codeB)
Introduction to Linear Regression Model (LRM) with spherical and non-spherical errors. Bayesian model comparison.
3.2 Multiple equation models (link to pdfC/codeC, link to pdfD/codeD)
LRM with Time-varying parameters and stochastic volatility. Multiple equation LRM (SUR, VAR, Panel models) with different prior distributions will be presented.
Assignment 4: (Text)
4 Bayesian Nonparametric Methods
4.1 Introduction to Bayesian Nonparametrics (BNP) (link to pdf/code)
Mixture Models, Infinite Mixture Models, Dirichlet and Pitman-Yor process priors, BNP clustering, Chinese Restaurant Processes, Slice sampling and MCMC sampling for nonparametric models.
4.2 BNP VAR models (link to pdf/code)
Nonparametric mixtures for array and BNP-VAR models, Dependent Pitman-Yor process priors, Application of BNP-VAR models to economics.
Assignment 5: (Text)
Many routines and functions used in the various parts of the course rely on material distributed by several scholars on their web pages or have been adapted from codes associated with Bayesian books or courses. In particular, some MATLAB functions come from LeSage's Econometrics Toolbox (see LeSage, Applied econometrics using MATLAB, available at http://www.spatial-econometrics.com). Some routines for VAR and hierarchical models are adapted from Koop (Bayesian Econometrics, Wiley, 2003, available at http://www.wiley.com/legacy/wileychi/koopbayesian/) and Koop and Korobilis (2010, material available at https://sites.google.com/site/garykoop/). Main codes usually acknowledge the sources.