Day 1 – Thursday, 9 Sept 2021

}

10:15 – 10:45

Opening Session

Opening Statements

}

11:00 - 12:15

Invited sessions #1

INV 1.1

Time series clustering

Organizer and Chair: Michele La Rocca, Pietro Coretto


Clustering financial time series using generalized cross correlations (p.27)
Andres M. Alonso, Carolina Gamboa and Daniel Peña

Network-based semi-supervised clustering of time series data (p.62)
Claudio Conversano, Giulia Contu, Luca Frigau and Carmela Cappelli 

Spatial-temporal clustering based on B-splines: robust models with applications to COVID-19 pandemic (p.83)
Pierpaolo D’Urso, Livia De Giovanni and Vincenzina Vitale

INV 1.2

Modern likelihood methods for model based-clustering

Organizer and Chair: Monia Ranalli

Discussant: Roberto Rocci


Hidden Markov and regime switching copula models for state allocation in multiple time-series (p.36)
Francesco Bartolucci, Fulvia Pennoni and Federico Cortese

Clustering data with non-ignorable missingness using semi-parametric mixture models (p.79)
Marie Du Roy de Chaumaray and Matthieu Marbac

Gaussian mixture models for high dimensional data using composite likelihood (p.98)
Michael Fop, Dimitris Karlis, Ioannis Kosmidis, Adrian O’Hagan, Caitriona Ryan and Isobel Claire Gormley

INV 1.3

Robust classification in action

Organizer and Chair: Marco Riani


Robust issues in estimating modes for multivariate torus data (p.21)
Claudio Agostinelli, Giovanni Saraceno and Luca Greco

Robust estimation of parsimonious finite mixture of Gaussian models (p.92)
Luis Angel García-Escudero, Agustín Mayo-Iscar and Marco Riani

Robust supervised clustering: some practical issues (p.142)
Fabrizio Laurini and Gianluca Morelli

INV 1.4

Flexible Bayesian mixture models for complex data

Organizer and Chair: Alessandra Guglielmi


Prediction of large observations via Bayesian inference for extreme-value theory (p.231)
Isadora Antoniano Villalobos, Simone Padoan and Boris Beranger

Model-based clustering for categorical data via Hamming distance (p.31)
Raffaele Argiento, Edoardo Filippi-Mazzola and Lucia Paci

MCMC computations for Bayesian mixture models using repulsive point processes (p.167)
Jesper Møller, Mario Beraha, Raffaele Argiento and Alessandra Guglielmi

}

12:30 – 13:30

KEY #1

Keynote #1 – Optimal transport methods for fairness in machine learning

Jean-Michel Loubes, Université Toulouse Paul Sabatier (FRANCE)

Chair: Francesca Grieselin

CLADAG 2021 Book of Abstracts and Short papers, p. 5 

}

13:30 – 14:00

Lunch break

}

14:00 – 15:00

Contributed sessions #1

CON 1.A

Models for clustering

Exploring solutions via monitoring for cluster weighted robust models (p.284)
Andrea Cappozzo, Luis Angel Garcìa Escudero, Francesca Greselin and Agustìn Mayo-Iscar

Clustering production indexes for construction with forecast distributions (p.360)
Michele La Rocca, Francesco Giordano and Cira Perna

Semi-constrained model-based clustering of mixed-type data using a composite likelihood approach (p.408)
Roberto Rocci and Monia Ranalli

Clustering models for three-way data (p.432)
Donatella Vicari and Paolo Giordani

 

CON 1.B

Nonparametric and semiparametric methods

Semiparametric finite mixture of regression models with Bayesian P-splines (p.268)
Marco Berrettini, Giuliano Galimberti and Saverio Ranciati

Angular halfspace depth: classification using spherical bagdistances (p.316)
Houyem Demni, Davide Buttarazzi, Stanislav Nagy and Giovanni Camillo Porzio

Functional cluster analysis of HDI evolution in European countries (p.336)
Francesca Fortuna, Alessia Naccarato and Silvia Terzi

A nonparametric test for mode significance (p.388)
Giovanna Menardi and Federico Ferraccioli

CON 1.C

Data analysis in biomedical science

A subject-specific measure of interrater agreement based on the homogeneity index (p.272)
Giuseppe Bove

Clustering longitudinal data with category theory for diabetic kidney disease (p.364)
Maria Mannone, Veronica Distefano, Claudio Silvestri and Irene Poli

Antibodies to SARS-CoV-2: an exploratory analysis carried out through the Bayesian profile regression (p.412)
Annalina Sarra, Adelia Evangelista, Tonio Di Battista and Damiana Pieragostino

Modelling three-way RNA sequencing data with mixture of multivariate Poisson-lognormal distribution (p.416)
Theresa Scharl and Bettina Grün

CON 1.D

Modelling dependence structures

An alternative to joint graphical lasso for learning multiple Gaussian graphical models (p.332)
Lorenzo Focardi Olmi and Anna Gottard

A Bayesian framework for structural learning of mixed graphical models (p.344)
Chiara Galimberti, Federico Castelletti and Stefano Peluso

Model selection procedure for mixture hidden Markov models (p.243)
Antonino Abbruzzo, Maria Francesca Cracolici and Furio Urso

CON 1.E

Data analysis in economics and finance

Predictive power of Bayesian CAR models on scale free networks: an application for credit risk (p.264)
Claudia Berloco, Raffaele Argiento and Silvia Montagna

A Machine Learning Approach in stock risk management (p.308)
Salvatore Cuomo, Federico Gatta, Fabio Giampaolo, Carmela Iorio and Francesco Piccialli

Clustering income data based on share densities (p.300)
Francesca Condino

Pathmox segmentation trees to compare linear regression models (p.312)
Cristina Davino and Giuseppe Lamberti

}

15:15 – 16:30

Invited sessions #2

INV 2.1

Copulas in time series analysis

Organizer and Chair: Marta L. Di Lascio, Roberta Pappadà


DTW-based assessment of the predictive power of the copula-DCC-GARCH-MST model developed for European insurance institutions (p.71)
Anna Denkowska and Stanisław Wanat

Nonlinear Interconnectedness of crude oil and financial markets (p.173)
Yarema Okhrin, Gazi Salah Uddin and Muhammad Yahya

Assessing food security issues in Italy: a quantile copula approach (p.195)
Giorgia Rivieccio, Jean-Paul Chavas, Giovanni De Luca, Salvatore Di Falco and Fabian Capitanio

INV 2.2

Advances in mixture models for matrix-variate and tensor data

Organizer and Chair: Antonio Punzo


Using Subset Log-Likelihoods to Trim Outliers in Gaussian Mixture Models
Katharine Clark and Paul McNicholas

Tensor-variate finite mixture model for the analysis of university professor remuneration (p.208)
Shuchismita Sarkar, Volodymyr Melnykov and Xuwen Zhu

Network analysis implementing a mixture distribution from Bayesian viewpoint (p.210)
Jarod Smith, Mohammad Arashi and Andriette Bekker

INV 2.3

Methods for inference from innovative or multiple data sources

Organizer and ChairEmilia Rocco, Chiara Bocci


Improving the reliability of a nonprobability web survey (p.120)
Yinxuan Huang and Natalie Shlomo

A nonparametric approach for statistical matching under informative sampling and nonresponse (p.146)
Daniela Marella and Danny Pfeffermann

Measurement errors in multiple systems estimation (p.211)
Paul Smith, Peter van der Heijden and Maarten Cruyff

INV 2.4

Social inequalities

Organizer and Chair: Mariangela Zenga


Socioeconomic inequalities and cancer risk: myth or reality? (p.106)
Carlotta Galeone

Quantifying the impact of covariates on the gender gap measurement: an analysis based on EU-SILC data from Poland and Italy (p.108)
Francesca Greselin and Alina Jędrzejczak

Gender inequalities from an income perspective (p.158)
Marcella Mazzoleni, Angiola Pollastri and Vanda Tulli

}

16:45 – 17:45

KEY #2

Keynote #2 – Veridical Data Science: the practice of responsible data analysis and decision-making

Bin Yu, University of California Berkeley (USA)

Chair: Maurizio Vichi

CLADAG 2021 Additional Abstracts

}

18:00 – 19:15

Invited sessions #3

INV 3.1

Recent advances in item response theory models

Organizer and Chair: Silvia Cagnone


Investigating model fit in item response models with the Hellinger distance (p.150)
Mariagiulia Matteucci and Stefania Mignani

Boosting multidimensional IRT models (p.40)
Michela Battauz and Paolo Vidoni

A study of lack-of-fit diagnostics for models fit to cross-classified binary variables (p.191)
Mark Reiser and Maduranga Dassanayake

INV 3.2

Advances in clustering

Organizer and Chair: Luca Frigau


Cluster validity by random forests (p.91)
Tahir Ekin and Claudio Conversano

Non-parametric consistency for the Gaussian mixture maximum likelihood estimator (p.116)
Christian Hennig and Pietro Coretto

Minimizing conflicts of interest: optimizing the JSM program (p.240)
Qiuyi Wu and David Banks

INV 3.3

Advances in parsimonious mixture modeling

Organizer: Volodymyr Melnykov

Chair: Xuwen Zhu


Parameter-wise co-clustering for high dimensional data (p.107)
Michael Gallaugher, Christophe Biernacki and Paul McNicholas

Transformation mixture modeling for skewed data groups with heavy tails and scatter (p.162)
Yana Melnykov, Xuwen Zhu and Volodymyr Melnykov

Clustering via new parsimonious mixtures of heavy tailed distributions (p.216)
Salvatore Daniele Tomarchio, Luca Bagnato and Antonio Punzo

INV 3.4

Bayesian non parametrics methods for classification

Organizer and Chair: Bruno Scarpa


Bayesian nonparametric dynamic modeling of psychological traits (p.25)
Emanuele Aliverti

Semiparametric IRT models for non-normal latent traits (p.178)
Sally Paganin

Malaria risk detection via mixed membership models (p.203)
Massimiliano Russo

Day 2 – Friday, 10 Sept 2021

}

9:15 - 10:30

CLADAG ASSEMBLY

Cladag assembly

}

10:45 – 12:00

Invited sessions #4

INV 4.1

Recent developments in symbolic data analysis

Organizer and ChairPaula Brito


A generalised clusterwise regression for distributional data (p.223)
Rosanna Verde, Francisco T. de A. De Carvalho and Antonio Balzanella

Detection of internet attacks with histogram principal component analysis (p.174)
M. Rosário Oliveira, Ana Subtil and Lina Oliveira

Identifying mortality patterns of main causes of death among young EU population using SDA approaches (p.141)
Simona Korenjak-Černe and Nataša Kejžar

INV 4.2

Recent advances in dynamic clustering: Markov models and extensions

Organizer and Chair: Daniele Tomarchio


Accounting for response behavior in longitudinal rating data (p.58)
Roberto Colombi, Sabrina Giordano and Maria Kateri

Modeling clusters of corporate defaults: regime-switching models significantly reduce the contagion source
Bård Støve, Geir D. Berentsen, Jan Bulla and Antonello Maruotti

Two-step estimation of multilevel latent class models with covariates (p.75)
Roberto Di Mari, Zsuzsa Bakk, Jennifer Oser and Jouni Kuha

INV 4.3

Networks data analysis and applications

Organizer and Chair: Mario R. Guarracino


Sender and receiver effects in latent space models for multiplex data (p.68)
Silvia D’Angelo

Networks of networks (p.186)
Panos Pardalos

Community detection in tripartite networks of university students mobility flows (p.232)
Maria Prosperina Vitale, Vincenzo Giuseppe Genova, Giuseppe Giordano and Giancarlo Ragozini

INV 4.4

Recent developments in the statistical analysis of categorical data

Organizer and ChairClaudia Tarantola


A risk indicator for categorical data (p.93)
Silvia Facchinetti and Silvia Angela Osmetti

A semi-Bayesian approach for the analysis of scale effects in ordinal regression models (p.124)
Maria Iannario and Claudia Tarantola

Simple effect measures for interpreting generalized binary regression models (p.129)
Maria Kateri

 

}

12:15 – 13:30

Invited sessions #5

INV 5.1

New issues in univariate and multivariate quantile regression

Organizer and Chair: Lea Petrella


Understanding and estimating conditional parametric quantile models (p.44)
Matteo Bottai

On model-based clustering using quantile regression (p.102)
Carlo Gaetan, Paolo Girardi and Victor Muthama Musau

Unconditional M-quantile regression (p.163)
Luca Merlo, Lea Petrella and Nikos Tzavidis

INV 5.2

Penalized techniques for data analysis

Organizer and Chair: Gianluca Sottile


Shapley Lorenz methods for eXplainable artificial intelligence (p.45)
Niklas Bussmann, Roman Enzmann, Paolo Giudici and Emanuela Raffinetti

Smoothed non linear PCA for multivariate data (p.54)
Marcello Chiodi

Causal regularization (p.236)
Ernst Wit and Lucas Kania

INV 5.3

Latent variable mixture modeling in epidemiology

Organizers: Maria Iannario


Characterising longitudinal trajectories of COVID-19 biomarkers within a latent class framework (p.64)
Federica Cugnata, Chiara Brombin, Pietro Cippà, Alessandro Ceschi, Paolo Ferrari and Clelia Di Serio

Pairwise likelihood estimation of latent autoregressive count models (p.187)
Xanthi Pedeli and Cristiano Varin

A machine learning approach for evaluating anxiety in neurosurgical patients during the COVID-19 pandemic (p.227)
Marika Vezzoli, Francesco Doglietto, Stefano Renzetti, Marco Fontanella and Stefano Calza

INV 5.4

Co-clustering for temporal sequences and distributional data

Organizer and Chair: Rosanna Verde


Mining multiple time sequences through co-clustering algorithms for distributional data (p.32)
Antonio Balzanella, Antonio Irpino and Francisco de A.T. De Carvalho

Co-clustering for high dimensional sparse data
Nicoleta Rogovschi

A general bi-clustering technique for functional data (p.217)
Agostino Torti, Marta Galvani, Alessandra Menafoglio, Piercesare Secchi and Simone Vantini

}

13:30 – 14:00

Lunch break

}

14:00 – 15:15

Invited sessions #6

INV 6.1

Advances in robust cluster analysis

Organizers: Luis Angel García-Escudero,  Agustín Mayo-Iscar, Marco Riani

Chair: Agustín Mayo-Iscar


Robust classification of spectroscopic data in agri-food: first analysis on the stability of results (p.49)
Andrea Cappozzo, Ludovic Duponchel, Francesca Greselin and Brendan Murphy

Robust classification in high dimensions using regularized covariance estimates (p.215)
Valentin Todorov and Peter Filzmoser

Issues in monitoring the EU trade of critical COVID-19 commodities(p.53)
Andrea Cerasa, Enrico Checchi, Domenico Perrotta and Francesca Torti

INV 6.2

Bayesian analysis of finite and infinite mixtures

Organizer and Chair: Bettina Grün


PIVMET: pivotal methods for Bayesian relabelling in finite mixture models (p.87)
Leonardo Egidi, Roberta Pappadà, Francesco Pauli and Nicola Torelli

A transdimensional MCMC sampler for spatially dependent mixture models (p.112)
Alessandra Guglielmi, Mario Beraha, Matteo Giannella, Matteo Pegoraro and Riccardo Peli 

Infinite mixtures of infinite factor analysers (p.168)
Keefe Murphy, Cinzia Viroli and Isobel Claire Gormley 

INV 6.3

Latent variable models for constructing composite indices

Organizers and Chairs: Rosaria Romano


PCA-based composite indices and measurement model (p.154)
Matteo Mazziotta and Adriano Pareto

Specifying composites in structural equation modeling: the Henseler-Ogasawara specification (p.209)
Florian Schuberth

Developing a multidimensional and hierarchical index following a composite-based approach (p.220)
Laura Trinchera

INV 6.4

Issues in directional data analysis

Organizer and Chair: Giovanni Camillo Porzio


Best approach direction for spherical random variables (p.128)
Jayant Jha

Angular halfspace depth: computation (p.169)
Stanislav Nagy, Petra Laketa and Rainer Dyckerhoff 

Nonparametric estimation of the number of clusters for directional data (p.207)
Paula Saavedra-Nieves and Rosa M. Crujeiras

}

15:30 – 16:30

KEY #3

Keynote #3 – Understanding cross-validation and prediction error

Robert Tibshirani, Stephen Bates and Trevor Hastie

Chair: Anna Gottard 

CLADAG 2021 Book of Abstracts and Short papers, p. 7

}

16:45 – 17:45

KEY #4

Keynote #4 – Class maps for visualizing classification results

Peter Rousseeuw, Jakob Raymaekers and Mia Hubert

Chair: Marco Riani 

CLADAG 2021 Book of Abstracts and Short papers, p. 6

}

18:00 - 20:00

PLENARY

Plenary – Statistical Issues in the COVID-19 Pandemic

Organizer and Chair: J. Sunil Rao

A simple correction for COVID-19 sampling bias

Daniel Diaz

CLADAG 2021 Book of Abstracts and Short papers, p. 14

A seat at the table: the key role of biostatistics and data science in the COVID-19 pandemic

Jeffrey S. Morris

CLADAG 2021 Book of Abstracts and Short papers, p. 15

Predictions, role of interventions and the crisis of virus in India: a data science call to arms

Bhramar Mukherjee

CLADAG 2021 Book of Abstracts and Short papers, p. 16

Bhramar Mukherjee

CLADAG 2021 Book of Abstract and Short papers, p. 16

Contributions of Israel’s CBS to rout COVID-19

Danny Pfeffermann

CLADAG 2021 Book of Abstracts and Short papers, p. 17

Day 3 – Saturday, 11 Sept 2021

}

10:00 – 11:15

Invited sessions #7 & Contributed sessions #2

INV 7.1

Recent advances in directional statistics

Organizer and ChairStefania Fensore, Agnese Panzera


Mixtures of Kato–Jones distributions on the circle, with an application to traffic count data (p.133)
Shogo Kato, Kota Nagasaki and Wataru Nakanishi

How to design a directional distribution (p.137)
John Kent

A graphical depth-based aid to detect deviation from unimodality on hyperspheres (p.182)
Giuseppe Pandolfo

INV 7.2

Recent developments in flexible regression – methods and software

Organizer and ChairMarco Geraci


Additive quantile regression via the qgam R package (p.97)
Matteo Fasiolo 

Additive Bayesian variable selection under censoring and misspecification

Javier Rubio

CON 2.A

Robust methods and data diagnostic

A combined test of the Benford hypothesis with anti-fraud applications (p.256)
Lucio Barabesi, Andrea Cerasa, Andrea Cerioli and Domenico Perrotta

Multivariate outlier detection for histogram-valued variables (p.384)
Ana Martins, Paula Brito, Sónia Dias and Peter Filzmoser

 Angular halfspace depth: Central regions (p.356)
Petra Laketa and Stanislav Nagy

Robustness methods for modelling count data with general dependence structures (p.396)
Marta Nai Ruscone and Dimitris Karlis

A robust quantile approach to ordinal trees (p.424)
Rosaria Simone, Cristina Davino, Domenico Vistocco and Gerhard Tutz

CON 2.B

Web mining and textual data analysis

Unabalanced classfication of electronic invoicing (p.260)
Chiara Bardelli

The use of multiple imputation techniques for social media data (p.372)
Paolo Mariani, Andrea Marletta and Matteo Locci

Visualizing cluster of words: a graphical approach to grammar acquisition (p.392)
Massimo Mucciardi, Giovanni Pirrotta, Andrea Briglia and Arnaud Sallaberry

Using eye-traking data to create a weighted dictionary for sentiment analysis: the eye dictionary (p.436)
Gianpaolo Zammarchi and Jaromir Antoch 

}

11:30 – 12:30

Contributed sessions #3

CON 3.A

Data analysis in biology and environmental sciences

Model-based clustering for estimating cetaceans site-fidelity and abundance (p.292)
Gianmarco Caruso, Greta Panunzi, Marco Mingione, Pierfrancesco Alaimo Di Loro, Stefano Moro, Edoardo Bompiani, Caterina Lanfredi, Daniela Silvia Pace, Luca Tardella and Giovanna Jona Lasinio

Prediction of gene expression from transcription factors affinities: an application of Bayesian non-linear modelling (p.376)
Federico Marotta, Paolo Provero and Silvia Montagna

High dimensional model-based clustering of European georeferenced vegetation plots (p.380)
Francesca Martella, Fabio Attorre, Michele De Sanctis and Giuliano Fanelli

Bayesian analysis of a water quality high-frequency time series through Markov switching autoregressive models (p.400)
Roberta Paroli, Luigi Spezia, Marc Stutter and Andy Vinten

CON 3.B

Process and service quality evaluation

Estimating latent linear correlations from fuzzy contingency tables (p.276)
Antonio Calcagnì

The LP data depth and its application to multivariate process control charts (p.352)
Carmela Iorio, Giuseppe Pandolfo, Michele Staiano, Massimo Aria and Roberta Siciliano

Detecting the effect of secondary school in higher education university choices (p.404)
Mariano Porcu, Isabella Sulis and Cristian Usala

 

CON 3.C

Mixture models and random effect models

Estimating Bayesian mixtures of finite mixtures with telescoping sampling (p.340)
Sylvia Frühwirth-Schnatter, Bettina Grün and Gertraud Malsiner-Walli

Measurement error models on spatial network lattices: car crashes in Leeds (p.348)
Andrea Gilardi, Riccardo Borgoni, Luca Presicce and Jorge Mateu

A redundancy analysis with multivariate random-coefficients linear models (p.368)
Laura Marcis, Maria Chiara Pagliarella and Renato Salvatore

Group-dependent finite mixture model (p.304)
Paula Costa Fontichiari, Miriam Giuliani, Raffaele Argiento and Lucia Paci

CON 3.D

Machine learning and statistical learning

Semi-supervised Learning through depth functions (p.255)
Simona Balzano, Mario Rosario Guarracino and Giovanni Camillo Porzio

Model-based clustering with sparse matrix mixture models (p.280)
Andrea Cappozzo, Alessandro Casa and Michael Fop

Neural networks for high cardinality categorical data (p.320)
Agostino Di Ciaccio

Stacking ensemble of Gaussian mixtures (p.420)
Luca Scrucca

CON 3.E

Hierarchical clustering and classification methods

A full mixture of experts model to classify constrained data (p.247)
Roberto Ascari and Sonia Migliorati

Categorical classifiers in multi-class classification problems (p.288)
Maurizio Carpita and Silvia Golia

Ali-Mikhail-Haq copula to detect low correlations in hierarchical clustering (p.324)
F. Marta L. Di Lascio, Andrea Menapace and Roberta Pappadà

Model-based clustering with parsimonious covariance structure (p.296)
Carlo Cavicchia, Maurizio Vichi and Giorgia Zaccaria

}

12:45 – 13:45

KEY #5

Keynote #5 – Quantile-based classification

Cinzia Viroli, University of Bologna (ITALY)

Chair: Brendan Murphy 

CLADAG 2021 Book of Abstracts and Short papers, p. 8

}

13:45 – 14:00

CLOSING STATEMENTS

Closing Statements