Beatrice (Bea) Franzolini

News, Upcoming Talks, Sessions and Conferences
I am thrilled and honored to have received the [2024 IMS New Researcher Travel Award] from the Institute of Mathematical Statistics!

I will talk about how to leverage Bayesian mixtures for longitudinal clustering and classification in
May 2024: [StaTalk2024@DiSIA], in Florence, Italy.
June 2024: [52nd Meeting of the Italian Statistical Society (SIS 2024)], in Bari, Italy.

I will talk about multivariate species sampling models (joint work with A. Lijoi, I. Prünster, and G. Rebaudo) in
June 2024: [4th Italian Meeting on Probability and Mathematical Statistics 2024], Rome, Italy.
July 2024: [Interpretable Inference via Principled BNP Approaches in Biomedical Research and Beyond], in Singapore, Republic of Singapore.

I will talk about conditional partial exchangeability (joint work with M. De Iorio) and dynamic clustering in network data (joint work with F. Gaffi) in
June 2024: [International Symposium on Nonparametric Statistics (ISNPS 2024)], in Braga, Portugal.
July 2024: [ISBA World Meeting 2024], in Venice, Italy.
August 2024: [Bernoulli-IMS 2024 World Congress] in Bochum, Germany.
September 2024: [Frontiers of Bayesian Inference and Data Science Workshop] in Oaxaca, Mexico. (I will join virtually)

I will talk about hierarchical multiplex networks (joint work with D. Durante and V. Ghidini) in
July/August 2024: [3rd BNP Networking Workshop], in Singapore, Republic of Singapore

Do not miss:
June 2024 the [Satellite workshop to International Society for Bayesian Analysis (ISBA) world meeting] in Lugano, Switzerland.
June 2024 the [Bayesian Young Statisticians Meeting (BAYSM)] in Venice, Italy.

Publications

Preprints and submitted articles
  • Franzolini, B., De Iorio, M., and Eriksson, J. (2023). Conditional partial exchangeability: a probabilistic framework for multi-view clustering. arXiv:2307.01152 [pdf] [code]

Articles in refereed journals
  1. Ascolani, F., Franzolini, B., Lijoi, A., and Prünster, I. (2024). Nonparametric priors with full-range borrowing of information. Biometrika, forthcoming. DOI:10.1093/biomet/asad063 [pdf] [code]
  2. Franzolini, B. and Rebaudo, G. (2024). Entropy regularization in probabilistic clustering. Statistical Methods & Applications, forthcoming. DOI:10.1007/s10260-023-00716-y [pdf] [code]
  3. Franzolini, B., Beskos, A., De Iorio, M., Poklewski Koziell, W., and Grzeszkiewicz, K. (2024). Change point detection in dynamic Gaussian graphical models: the impact of COVID-19 pandemic on the US stock market. The Annals of Applied Statistics, 18(1), 555-584. DOI:10.1214/23-AOAS1801 [pdf] [code]
  4. Franzolini, B., Cremaschi, A., van den Boom, W., and De Iorio, M. (2023). Bayesian clustering of multiple zero-inflated outcomes. Philosophical Transactions of the Royal Society A, 381(2247): 20220145. DOI:10.1098/rsta.2022.0145 [pdf] [code]
  5. Franzolini, B., Lijoi, A., and Prünster, I. (2023). Model selection for maternal hypertensive disorders with symmetric hierarchical Dirichlet processes. The Annals of Applied Statistics 17(1): 313-332. DOI:10.1214/22-AOAS1628 [pdf] [code]

Refereed discussions and conference proceedings
  1. Franzolini, B. (2024). How to leverage Bayesian mixtures for dynamic clustering and classification In Book of Short Papers SIS 2024, in press. [pdf] [code]
  2. Bondi, L., Franzolini, B., Palma, M. (2024). A longitudinal study of the gender gap in school grades via flexible Bayesian Beta regression In Book of Short Papers SIS 2024, in press. [pdf]
  3. Franzolini, B., Bondi, L., Fasano, A., and Rebaudo, G. (2023). Bayesian forecasting of multivariate longitudinal zero-inflated counts: an application to civil conflict. In Book of Short Papers CLADAG 2023, 465-468. (Editors: Corretto, P., Giordano G., La Rocca, M., Parrella, M. L., Rampichini, C.) ISBN 9788891935632 [book] [pdf]
  4. Fasano, A., Anceschi, N., Franzolini, B., and Rebaudo, G. (2023). Efficient computation of predictive probabilities in probit models via expectation propagation. In Book of Short Papers CLADAG 2023, 449-452. (Editors: Corretto, P., Giordano G., La Rocca, M., Parrella, M. L., Rampichini, C.) ISBN 9788891935632 [book] [pdf] [code]
  5. Fasano, A., Anceschi, N., Franzolini, B., and Rebaudo, G. (2023). Efficient expectation propagation for posterior approximation in high-dimensional probit models. In Book of Short Papers SIS 2023, 1133-1138. (Editors: Chelli, F. M., Ciommi, M., Mariani, F., Recchioni, M. C.) ISBN 9788891935618 [book] [pdf] [code]
  6. Rebaudo G., Fasano, A., Franzolini, B., and Müller, P. (2023) A discussion on: “Evaluating sensitivity to the stick-breaking prior in Bayesian nonparametrics” by Giordano, R., Liu, R., Jordan M. I. and Broderick T. Bayesian Analysis, 18(1): 287-366. DOI: 10.1214/22-BA1309 [pdf]
  7. Franzolini, B. and Rebaudo, G. (2022). A regularized-entropy estimator to enhance cluster interpretability in Bayesian nonparametrics. In Book of Short Papers SIS 2022, 387-397. (Editors: Balzanella, A., Bini, M., Cavicchia, C. and Verde, R.) ISBN 9788891932310 [book] [pdf]
  8. Ascolani, F., Franzolini, B., Lijoi, A., and Prünster, I. (2021). On the dependence structure in Bayesian nonparametric priors. In Book of Short Papers SIS 2021, 1219-1225. (Editors: Perna, C., Salvati, N. and Schirripa Spagnolo, F.) ISBN 9788891927361 [book-part1] [book-part2] [pdf]

PhD Thesis
  • Franzolini, B. (Advisors: Lijoi, A., and Prünster, I.), Feb 2022. On Dependent Processes in Bayesian Nonparametrics: Theory, Methods, and Applications. Bocconi University [pdf - high quality] [pdf - compressed]

Projects

Ongoing research projects
Multivariate species sampling models (joint work with A. Lijoi, I. Prünster and G. Rebaudo, in preparation)
- a preliminary version can be found in Chapter 2 of the [PhD thesis] )

Geometry of Scale mixtures (joint work with M. De Iorio, in preparation)

Hierarchical multiplex networks (joint work with D. Durante and V. Ghidini, in preparation)

An invariance-based approached to community detection in dynamic networks (joint work with F. Gaffi, in preparation)

Scalable expectation propagation for generalized linear models (joint work with N. Anceschi, A. Fasano and G. Rebaudo, in preparation)

Dependent prior processes for panel count data (joint work with A. Lijoi and I. Prünster, in preparation)
- a preliminary version can be found in Chapter 5 of the [PhD thesis] )


Teaching

A.Y. 2023-2024
BOCCONI UNIVERSITY
Bayesian Statistical Methods - Course Director
A.Y. 2020-2021
BOCCONI UNIVERSITY
Statistica (BSc in Economics and Management) - Adjunct Professor
Financial Econometrics and Empirical Finance (MSc in Finance) - Adjunct Professor
A.Y. 2019-2020
BOCCONI UNIVERSITY
Statistica (BSc in Economics and Management) - Adjunct Professor
Financial Econometrics and Empirical Finance (MSc in Finance) - Teaching Assistant
UNIVERSITÀ CATTOLICA DEL SACRO CUORE
Statistics (BSc in Economics and Management) - Teaching Assistant
A.Y. 2018-2019
BOCCONI UNIVERSITY
Statistica (BSc in Economics and Management) - Teaching Assistant
Financial Econometrics and Empirical Finance (MSc in Finance) - Teaching Assistant
Metodi Quantitativi per la Finanza (BSc in Economics and Finance) - Teaching Assistant

About me

A full pdf version of my CV is available here.
I am a Researcher (Italian RTD-A) at the Bocconi Institute for Data Science and Analytics , at Bocconi University, Italy.

My research interests are in Bayesian Nonparametrics, Probabilistic Machine Learning, Applied Bayesian Modelling and Computational Statistics with a focus on
dependent random measures,
clustering, random partition models and network data
change-point detection and dynamic models,
biostatistics, behavioural and financial econometrics applications.

I am the Chair of the Junior Section of ISBA (j-ISBA) .


Until July 2023, I was a Research Fellow at the Agency for Science, Technology and Research (A*STAR) and member of the Division of Biomedical Data Science (BiDS), led by Prof. Maria De Iorio, at the National University of Singapore (NUS), in Singapore. Until January 2022, I was a PhD student in Statistics at Bocconi University, Italy, under the supervision of Prof. Antonio Lijoi and Prof. Igor Prünster.