Three independent tutorials will take place prior to the conference.

Dates: 13-15 December 2023.
Venue: HTW Berlin, University of Applied Sciences (Wilhelminenhof campus).
Room: Rooms 007, ground floor, Building G. The coffee breaks will take place at Room 008 (download details here). For virtual access, see below.

The tutorials are organized by the COST Action HiTEc that offers the possibility of attending to the 3 tutorials independently from the conference (see HiTEc Winter Course 2023). HiTEc also offers the possibility of applying for grants (see below). The conference participants can register for each one of the tutorials separately. For further information send an email to info@CMStatistics.org.

A link with some material will be provided to the students in due course.

Bayesian semiparametric regression.

Presenters: Thomas Kneib, Hannes Riebl, Paul Wiemann.
Email: Contact

Dates: December 13th and 14th morning, 2023.

Semiparametric regression models overcome some of the restrictions of classical forms of regression models such as (i) the linearity of covariate effects, (ii) the independence of observations, or (iii) the focus on specific types of response distributions and on modelling the conditional mean alone. In this sense, semiparametric regression forms an overarching model class comprising various special cases such as generalized additive models (GAMs), models with random effects, spatial regression models, or generalized additive models for location, scale, and shape (GAMLSS). Bayesian inference based on Markov chain Monte Carlo (MCMC) simulation techniques provides a particularly attractive way of statistically treating such models and developing extensions. This is, for example, due to the modularity of MCMC that allows to flexibly combine different blocks and algorithms for different types of model parameters.

This tutorial will combine lectures with practical exercises on the implementation of Bayesian semiparametric regression utilizing the novel probabilistic programming environment Liesel (https://liesel-project.org). Liesel is developed with the aim of supporting research on Bayesian inference based on MCMC simulation in general and semiparametric regression in particular. Its three main components are (i) an R interface (RLiesel) for the configuration of initial semiparametric regression models, (ii) a graph-based model-building library where the initial model graph can be manipulated to incorporate new research ideas, and (iii) an MCMC library for designing modular inference algorithms combining multiple types of well-tested MCMC kernels.

In the tutorial, we will build on Liesel and discuss (i) general principles of Bayesian inference with MCMC, (ii) Bayesian additive models, and (iii) Bayesian distributional regression. We will combine theoretical background information with hands-on work on applications for all course parts. For participating, knowledge of the principles of Bayesian inference, familiarity with linear and generalized models, and some experience in statistical programming with Python or R will be beneficial.

The instructors for the tutorial will be Thomas Kneib (University of Göttingen), Paul Wiemann (Texas A&M University), and Hannes Riebl (University of Göttingen). Thomas Kneib is a Professor of Statistics and has contributed to the field of Bayesian semiparametric regression with new statistical methodology, as well as the development of software and applications in various contexts. Paul Wiemann and Hannes Riebl are postdoctoral researchers and Liesel’s leading developers.

Risk management with vine copula based dependence models.

Prof. Claudia Czado, Oezge Sahin, Karoline Bax, Technical University of Munich, Germany.
Email: Contact

Dates: December 14th afternoon and 15th morning, 2023.

In the complex risk management landscape, exploring dependency structures becomes paramount. Copulas are key tools in such exploration. However, since standard copula models do not provide flexible dependence and tail patterns, vine copula models (vine-copula.org) were designed to increase the flexibility of these models and overcome their limitations, such as allowing for asymmetric tail dependence. Delving into vine copula-based dependence models, we introduce the fundamental concepts of copulas and vine copulas in the first part of the tutorial. This foundation aids in understanding and using the diverse R software tools available for vine copula analysis, like rvinecopulib of Nagler and Vatter (2023). Progressing further in the second part, we will explore how vine copulas can be used to give insight into multivariate time series data. In this context, we will discuss the integration of univariate ARMA-GARCH models with vines and explore how vine copula models can change how we manage risk. Further, we will look at stress testing using vine copulas. We will provide an overview of R libraries useful in the analysis of multivariate time series: the portvine package of Sommer (2023) is key for analyzing portfolios, especially when Expected Shortfall (ES) and Value at Risk (VaR) are to be estimated. Additionally, we will look at stress testing using D-vine copula-based regressions implemented in the vinereg of Nagler and Kraus (2022) package.

Using real financial data, we will show an implementation, and participants will have the opportunity to get hands-on experience. The tutorial will offer a combination of lectures with practical exercises using the R software.

The instructors for the tutorial will be Claudia Czado (Technical University of Munich), Özge Şahin (Delft University of Technology), and Karoline Bax (Technical University of Munich). Claudia Czado is Associate Professor of Applied Mathematical Statistics and has contributed to the field of copulas and vine copulas. Özge Şahin is an Assistant Professor working on statistical learning and dependence models. Karoline Bax is a Postdoctoral Researcher focusing on Sustainable Finance.

References:

Czado, C., & Nagler, T. (2022). Vine copula based modeling. Annual Review of Statistics and Its Application, 9, 453-477.

Czado, C., Bax, K., Sahin, Ö., Nagler, T., Min, A., & Paterlini, S. (2022). Vine copula based dependence modeling in sustainable finance. The Journal of Finance and Data Science.

Sommer, E., Bax, K., & Czado, C. (2022). Vine Copula based portfolio level conditional risk measure forecasting. arXiv preprint arXiv:2208.09156. Accepted for publication in CSDA.

Network econometrics.

Presenters: Monica Billio, Roberto Casarin
Email: Contact

Dates: December 15th afternoon 2023.

The tutorial aims to provide students with models and tools from graph theory to extract graphs from time series and illustrate the recent advances in measuring financial interconnectedness [1].

The organization of the course can be split into two main parts. The first introduces some background in extracting latent networks with single-layer and multiple-layer Graphical Vector Autoregressive (GVAR) models [2,3,4]. In GVAR models, the structural VAR model's contemporaneous and temporal causal structures are represented by two different graphs. A Bayesian approach and an efficient Markov chain Monte Carlo algorithm allow for estimating jointly the two causal structures and the parameters of the reduced form VAR model.

In the second part, the two graphs are used to measure of two types of connectedness effects in two applications. The first application is to the stock market [3] and focuses on the linkages between financial and non-financial super-sectors. The second application considers linkages in the oil markets following two transmission layers: production and rigs [5]. Participants will learn to deal with network analysis in MATLAB and Gephi.

The instructor for the tutorial will be Roberto Casarin. He is a professor of econometrics at Ca' Foscari University of Venice and director of the Venice Center for Risk Analytics. His research is mainly in Bayesian nonparametric and semiparametric, Monte Carlo methods and time series models.

References:

[1] Billio M., Getmanki M., Lo A., Pelizzon L. Econometric Measures of Connectedness and Systemic Risk in the Finance and Insurance Sectors in Journal of Finacial Economics, vol. 104, pp. 535-559

[2] Ahelegbey D. F., Billio, M. and Casarin, R. (2016), Bayesian Graphical Models for Structural Vector Autoregressive Processes, Journal of Applied Econometrics, 31(2), 357-386.

[3] Ahelegbey D. F., Billio, M., Casarin, R. (2016), Sparse Graphical Multivariate Autoregression: A Bayesian approach, Annals of Economics and Statistics,123/124, 1-30.

[4] Bianchi, D., Billio, M., Casarin, R., Guidolin, M. (2018), Modeling Systemic Risk with Markov Switching Graphical SUR Models, Journal of Econometrics, 210(1), 58-74.

[5] Casarin, R., Iacopini, M., German, M., Ter Horst, E., Espinasa, R., Sucre, C. and Rigobon, R. (2020), Multilayer network analysis of oil linkages, Econometrics Journal, 23(2), 269–296.

Organized by the HiTEc COST Action CA21163 with the collaboration of CFE-CMStatistics.

Sponsored by COST.