Day 1 – Wednesday Sept, 11



Palazzo degli studi, Folcara

Registration open


9:15 – 9:30

Main Hall

Opening Statements

Giovanni Betta (Rector, University of Cassino) | Enrica Iannucci Head of the Economics and Law Department, Univesity of Cassino) | Roberto Rocci (Cladag President) | Francesca Greselin (Chair, Scientific Committee)


9:30 – 10:45

Invited sessions #1

Room 1

Archetypal analysis and related methods

Organizer and Chair: Domenico Vistocco (Università di Napoli Federico II, ITALY)

Archetypal contour shapes
Aleix Alcacer (Universitat Jaume I, SPAIN)Irene Epifanio López (Universitat Jaume I, SPAIN)M. Victoria Ibáñez (Universitat Jaume I, SPAIN)Amelia Simó (Universitat Jaume I, SPAIN)

Archetypes, prototypes and other types
Francesco Palumbo (Università di Napoli Federicco II, ITALY)

Probabilistic archetypal analysis
Sohan Seth (University of Edinburgh, SCOTLAND)

Room 2

Clustering time dependent data

Organizer and Chair: Antonello Maruotti (Università di Roma LUMSA, ITALY)

Model-based clustering of time series data: a flexible approach using nonparametric state-switching quantile regression models
Timo Adam (Bielefeld University, GERMANY)Roland Langrock (Bielefeld University, GERMANY)Thomas Kneib (University of Göttingen, GERMANY)

Multivariate change-point analysis for climate time series
Gianluca Mastroantonio (Politecnico di Torino, ITALY)

How to measure material deprivation? A Latent Markov Model based approach
Francesco Dotto (Università di Roma Tre, ITALY)

Room 3

New classification methods in Economics and Business

Organizer and Chair: Józef Pociecha (Uniwersytet Ekonomiczny w Krakowie, POLAND)

Comparison of serious diseases mortality in regions of V4
Lucie Kopecká (University of Pardubice, CZECH REPUBLIC) – Viera Pacakova (University of Pardubice, CZECH REPUBLIC)

Application of data mining in the housing affordability analysis
Viera Labudová (Bratislava University of Economics, SLOVAKIA)
Lubica Sipkova (Bratislava University of Economics, SLOVAKIA)

Ontology-based classification of multilingual corpuses of documents
Paweł Lula (Cracow University of Economics, POLAND)
S. Belov (Plechanov Russian University of Economics, Moscow, RUSSIAN FEDERATION) – Salvatore Ingrassia (Università di Catania, ITALY) – Zoran Kalinic (University of Kragujevac, SERBIA)

Main Hall

Recent methodological advances in finite mixture modeling with applications

Organizer and Chair: Volodymyr Melnykov (University of Alabama, US)

Skewed Distributions or Transformations? Incorporating Skewness in a Cluster Analysis
Michael Gallaugher (McMaster University, CANADA)Paul McNicholas (McMaster University, CANADA)Volodymyr Melnykov (University of Alabama, US)Xuwen Zhu (University of Louiseville, US)

On a Class of Repulsive Mixture Models
Jose Quinlan (Pontificia Universidad Catolica de Chile, CHILE)
Garritt Page (Brigham Young University, US) – Fernando Quintana (ontificia Universidad Catolica de Chile, CHILE)

Bayesian clustering using non-negative matrix factorization
Michael Porter (University of Virginia, US)
Ketong Wang (Amazon, US)


10:45 – 11:05

Coffee break


11:05 – 12:05

Main Hall

Keynote #1 – Statistics with a human face

Adrian Bowman, University of Glasgow (SCOTLAND)


Three-dimensional surface imaging, through laser-scanning or stereo-photogrammetry, provides high-resolution data defining the surface shape of objects.  Human faces are of particular interest and there are many biological and anatomical applications, including assessing the success of facial surgery and investigating the possible developmental origins of some adult conditions.  An initial challenge is to structure the raw images by identifying features of the face.  Ridge and valley curves provide a very good intermediate level at which to approach this, as these provide a good compromise between informative representations of shape and simplicity of structure.  Some of the issues involved in analysing data of this type will be discussed and illustrated.  Modelling issues include simple comparison of groups, the measurement of asymmetry and longitudinal patterns of shape change.  This last topic is relevant at short scale in facial animation, medium scale in individual growth patterns, and very long scale in phylogenetic studies.


12:05 – 13:20

Invited sessions #2

Main Hall

GfKl - Applications and Concepts of Classification

Organizer and Chair: Adalbert Wilhelm (Jacobs University, GERMANY)

Symmetry in Graph Clustering
Andreas Geyer-Schulz (Karlsruhe Institute of Technology, GERMANY)Fabian Ball (Karlsruhe Institute of Technology, GERMANY)

Invariant concept classes for transcriptome classification
Hans A. Kestler (Universität Ulm, GERMANY)
Ludwig Lausser (Universität Ulm, GERMANY) – Robin Szekely (Universität Ulm, GERMANY) – Attila Klimmek (Universität Ulm, GERMANY)

Price and product design strategies for manufacturers of electric vehicle batteries: inferences from latent class analysis
Friederike Paetz (Technical University of Clausthal, GERMANY)

Room 1

Latent variable models for complex networks - I

Organizers: Marco Alfò, Università di Roma “La Sapienza”, ITALY – Francesco Bartolucci, Università di Perugia, ITALY

A latent space model for clustering in multiplex data
Silvia D’Angelo (University College Dublin, IRELAND)Michael Fop (University College Dublin, IRELAND)

Clustering two-mode binary network data with overlapping mixture model and covariates information
Saverio Ranciati (Università di Bologna, ITALY)
Veronica Vinciotti (Brunel University, UK) – Ernst Wit (Università della Svizzera Italiana, SWITZERLAND) – Giuliano Galimberti (Università di Bologna, ITALY)

Multilinear tests of association between networks
Daniel K Sewell (The Iowa State University, US)

Room 2

Multi-block data analysis: methods and applications

Organizer and Chair: Rosaria Romano (Università di Napoli Federico II, ITALY)

Tensor analysis can give better insight
Rasmus Bro (University of Copenhagen, DENMARK)

Quantile composite-based path modeling to estimate the conditional quantiles of health indicators
Pasquale Dolce (Università di Napoli Federico II, ITALY) – Cristina Davino (Università di Napoli Federico II, ITALY)
Stefania Taralli (ISTAT, ITALY) – Domenico Vistocco (Università di Napoli Federico II, ITALY)

A canonical representation for multiblock methods
Mohamed Hanafi (Oniris, StatSC, Nantes, FRANCE)

Room 3

Statistical data depths

Organizer and Chair: Claudio Agostinelli (Università di Trento, ITALY)

A generalization of multivariate depth functions
Giacomo Francisci (Università di Trento, ITALY)Claudio Agostinelli (Università di Trento, ITALY)Alicia Nieto Reyes (Universidad de Cantabria, SPAIN)

Illumination in depth analysis
Stanislav Nagy (Charles University in Prague, CZECH REPUBLIC)
Jiří Dvořák (Charles University in Prague, CZECH REPUBLIC)

Depths for scatter and shape parameters
Germain Van Bever (Université de Namur, BELGIUM)
Davy Paindaveine (Université Libre de Bruxelles, BELGIUM)


13:20 – 14:20



14:20 – 15:10

Contributed sessions #1

Room 1

Data modelling and its applications

Generalizing the skew-t model using copulas
Antonio Parisi (Università di Roma Tor Vergata, ITALY)Brunero Liseo (Università di Roma “La Sapienza”, ITALY)

Decomposition of the interval based composite indicators by means of biclustering
Carlo Drago (University “Niccolò Cusano” in Roma, ITALY)

Sampling properties of an ordinal measure of interrater absolute agreement
Giuseppe Bove (Università Roma Tre, ITALY)
Pier Luigi Conti (Università “La Sapienza”, ITALY) – Daniela Marella (Università Roma Tre, ITALY)

Main Hall

Distances for complex data and Benford's law

A Mahalanobis-like distance for cylindrical data
Lucio Palazzo (Università di Cassino e del Lazio Meridionale, ITALY) – Giovanni C. Porzio (Università di Cassino e del Lazio Meridionale, ITALY) – Giuseppe Pandolfo (Università di Napoli Federico II, ITALY)

Detecting and interpreting the consensus ranking based on the weighted Kemeny distance
Alesso Baldassarre (Università di Cagliari, ITALY)
Claudio Conversano (Università di Cagliari, ITALY) – Antonio D’Ambrosio (Università di Napoli Federico II, ITALY)

Euclidean distance as a measure of conformity to Benfordʼs law in digital analysis for fraud detection
Mateusz Baryła (Cracow University of Economics, POLAND)
Józef Pociecha (Cracow University of Economics, POLAND)

Room 2

Regression methods and data reduction

Dynamic modelling of price expectations
Rosaria Simone (Università di Napoli Federico II, ITALY)Domenico Piccolo (Università di Napoli Federico II, ITALY) – Marcella Corduas (Università di Napoli Federico II, ITALY)

Marginal effects for comparing groups in regression models for ordinal outcome when uncertainty is present
Maria Iannario (Università di Napoli Federico II, ITALY)
Claudia Tarantola (Università di Pavia,ITALY)

A new approach to preference mapping through quantile regression
Cristina Davino (Università di Napoli Federico II, ITALY) – Rosaria Romano (Università di Napoli Federico II, ITALY) – Tormod Naes (NOFIMA, NORWAY) – Domenico Vistocco (Università di Napoli Federico II, ITALY)

Room 3

Statistical learning and data mining

Classify x-ray images using convolutional neural networks
Federica Crobu (Università di Roma “La Sapienza, ITALY) – Agostino di Ciaccio (Università di Roma “La Sapienza, ITALY)

Hidden Markov Models for clustering functional data
Andrea Martino (Politecnico di Milano, ITALY) – Giuseppina Guatteri (Politecnico di Milano, ITALY) – Anna Maria Paganoni (Politecnico di Milano, ITALY)

Sparse linear regression via random projections ensembles
Laura Anderlucci (Università di Bologna, ITALY)
Matteo Farnè (Università di Bologna, ITALY) – Giuliano Galimberti (Università di Bologna, ITALY) – Angela Montanari (Università di Bologna, ITALY)


15:10 - 16:40

Main Hall

Round table: Artificial intelligence and big data in business

Organizer and moderator: Vincenzo Esposito Vinzi, ESSEC Business School

AI at the heart of outstanding customer experiences

 The future is now and it is all about customer experiences.  But what does a beautiful experience look like?  We explore how Artificial Intelligence (AI) is enabling a new era of personalization driven by deep insight and predictive optimization.  We will look at how AI enables organizations to sense, comprehend, act and learn, and review how Accenture has put AI at the heart of its role as Innovation Partner at Carnival Corporation to transform customer experiences.

Jean-Pierre Bokobza: Senior Managing Director – Global Geographic Sales Lead, Accenture Digital
How AI & Big Data technology are changing the customer expectation, offering new ways to offer broader and personalized experiences?

Lots of industries are facing the challenges of digital transformation, customer have more choices and are now caring about experience. We will show, through real use cases, how we can leverage the advances in AI and predictive analytics to create unique offers and experiences at scale, that appeal to the needs and desires of each individual customer.

Fikria Chaouki: Senior expert Advanced analytics & AI for business – Former Vice President of Advanced analytics & AI at Accor
Data to change: knowledge sharing between research and firms in the statistical analysis of Big data

Technological evolution in the production of data does not always go hand in hand with a cultural evolution, which enables non-specialists to understand and take advantage of innovation. The cultural evolution to understand the key points and meaning of the profiling of the consumption behaviors, reading habits, and attitudes in real time is deeply rooted into statistics and data analysis.

Data and statistical literacy have particular relevance in the activities of companies and firms, when complex analyses are undertaken by teams from diverse backgrounds.

The question on how academics, statistical associations and firms can cooperate to promote and develop statistical literacy to enhance not only change and innovation, but also progress in societies has not yet a clear answer.

Monica Pratesi: SIS President – Full Professor of Statistics, University of Pisa
How are companies using big data today? Some concrete examples.

The private industry is currently putting a lot of efforts to transform their businesses into smart businesses. In fact, the full digitalisation of companies activities’ generates data usable for increasing operational efficiency and for creating new products and services. The characteristics of this data necessitates automated procedure typically using AI. Based on strategic big data analytics case studies from large companies, we give concrete examples of the type of data used, the algorithms and the data valorisation.

Jeroen Rombouts: Head of Information Systems, Decision Sciences and Statistics Department, Accenture Strategic Business Analytics Chair – ESSEC Business School
Mobile Data for Social Good: How Mobile Operators are Building Sustainable and Responsible Business.

ICTs are an important enabler of economic growth and development. Mobile phones technology is experiencing massive penetration rates globally, even in low and middle-income locations, reaching urban and rural populations, across all socio-economic spectra. Each mobile device generates an incredible amount of information that yields unique insights on human behaviour, particularly relevant and promising for the social sector. In 2015, United Nations have called for a ‘data revolution’, as an enabler of evidence-based policies and programs to reach the most vulnerable. Mobile operators are working closely with public agencies and NGOs to establish a common framework and ecosystem that can help provide that information, to eventually improve the lives of billions of people and the environment in which they live. Orange has also supported this cause, launching several initiatives and studies that have shown the value of engaging through and about data.

Stefania Rubrichi: Research scientist – ORANGE’s SENSE (Sociology and Economics of Networks and Services) Laboratory


16:40 – 16:55

Coffee break


16:55 – 18:10

Invited sessions #3

Room 1

Advances in recursive partitioning (with applications)

Organizer and Chair: Claudio Conversano (Università di Cagliari, ITALY)

Flexible model-based trees for count data
Federico Banchelli (Università di Bologna, ITALY)

Modeling Heterogeneity in Clustered Data using Recursive Partitioning
Moritz Berger (University of Bonn, GERMANY)
Gerhard Tutz (Ludwig Maximilian University of Munich, GERMANY)

Tree modeling ordinal responses: CUBREMOT and its applications
Carmela Cappelli (Università di Napoli Federico II, ITALY)
Rosaria Simone (Università di Napoli Federico II, ITALY) – Francesca Di Iorio (Università di Napoli Federico II, ITALY)

Room 2

Challenges in modern robust data analysis

Organizer and Chair: Andrea Cerioli (Università di Parma, ITALY)

Robust model-based clustering with mild and gross outliers
Alessio Farcomeni (Università di Roma “La Sapienza”, ITALY)Antonio Punzo (Università di Catania, ITALY)

Issues in nonlinear time series modeling of European import volumes
Gianluca Morelli (Università di Parma, ITALY)
Francesca Torti (Joint Research Centre of the European Commission, ITALY)

Multivariate outlier detection in high reliability standards fields using ICS
Anne Ruiz-Gazen (Toulouse School of Economics, FRANCE)
Aurore Archimbaud (Ippon Innovation, FRANCE) – Klaus Nordhausen (Vienna University of Technology, AUSTRIA)

Room 3

Cluster validation and the selection of a cluster configuration

Organizer and Chair: Pietro Coretto (Università di Salerno, ITALY)

Likelihood – type methods for comparing clustering solutions
Luca Coraggio (Università di Napoli Federico II, ITALY)Pietro Coretto  (Università di Salerno, ITALY)

An adequacy approach to estimating the number of clusters
Christian Hennig (Università di Bologna, ITALY)

New developments in application of pairwise overlap
Volodymyr Melnykov (University of Alabama, US)
Yana Melnykov (University of Alabama, US) – Domenico Perrotta (European Commission, Joint Research Centre, ITALY) – Marco Riani (Università di Parma, ITALY) – Francesca Torti (European Commission, Joint Research Centre, ITALY) – Yang Wang (University of Alabama, US)

Main Hall

Generalized Linear, Latent, and Mixed Models

Organizer and Chair: Matilde Bini (Università Europea di Roma, ITALY)

Some issues in generalized linear modeling
Alan Agresti, (University of Florida, US)

Evaluating the school effect: adjusting for pre-test or using gain scores?
Bruno Arpino (Università di Firenze, ITALY)
Silvia Bacci (Università di Firenze, ITALY) – Leonardo Grilli (Università di Firenze, ITALY) – Leonardo Grilli (Università di Firenze, ITALY) – Carla Rampichini (Università di Firenze, ITALY)

Statistical analysis of item pre-knowledge in educational test: latent variable modelling and optimal statistical decision
Chen Yunxiao (London School of Economics, UK)
Lu Yan (London School of Economics, UK) – Irini Moustaki (London School of Economics, UK)


18:10 – 19:10

Room 1

Cladag assembly



Main Hall

MLT Concert



Networking evening with food and drinks

(included in the conference fees)

Day 2 – Thursday Sept, 12


8:00 – 8:50

Contributed sessions #2

Main Hall

Applications in finance and insurance

A multi-criteria approach in a financial portfolio selection framework
Carmela Iorio (Università di Napoli Federico II, ITALY) – Giuseppe Pandolfo (Università di Napoli Federico II, ITALY) – Roberta Siciliano (Università di Napoli Federico II, ITALY)

Silhouette-based method for portfolio selection
Marco Scaglione (Università di Napoli Federico II, ITALY) – Carmela Iorio (Università di Napoli Federico II, ITALY) – Antonio D’Ambrosio (Università di Napoli Federico II, ITALY)

Application of survival analysis to critical illness insurance data
David Zapletal (University of Pardubice, CZECH REPUBLIC)
Lucie Kopecká (University of Pardubice, CZECH REPUBLIC)

Room 1

Clustering, classification and data analysis

Chair: TBA

AUC-based gradient boosting for imbalanced classification
Martina Dossi (European Central Bank, ITALY)Giovanna Menardi (IUniversità di Padova, ITALY)

Consensus clustering via pivotal methods
Leonardo Egidi (Università di Trieste, ITALY)
Roberta Pappadà (Università di Trieste, ITALY) – Francesco Pauli (Università di Trieste, ITALY) – Nicola Torelli (Università di Trieste, ITALY)

ROC curves with binary multivariate data
Lidia Sacchetto (Politecnico di Torino, ITALY)
Mauro Gasparini (Politecnico di Torino, ITALY)

Room 2

Robustness and stability

On the robustness of the cosine distribution depth classifier
Houyem Demni (Università di Cassino e del Lazio Meridionale, ITALY) – Amor Messaoud  (University of Tunis, TUNISIA) – Giovanni C. Porzio (Università di Cassino e del Lazio Meridionale, ITALY)


Stability of joint dimension reduction and clustering
Angelos Markos (Democritus University of Thrace, GREECE) – Michel Van de Velden (Erasmus University of Rotterdam, NETHERLANDS) – Alfonso Iodice D’Enza (Università di Napoli Federico II, ITALY)

Room 3

Supervised and unsupervised classification for complex data

A statistical model for software releases complexity prediction
Marco Ortu (Università di Cagliari, ITALY)
Giuseppe Destefanis (Brunel University, UK) – Roberto Tonelli (Università di Cagliari, ITALY)

Random projections of variables and units
Laura Anderlucci (Università di Bologna, ITALY)
Roberta Falcone (Università di Bologna, ITALY) – Angela Montanari (Università di Bologna, ITALY)


8:50 – 10:05

Invited sessions #4

Room 1

Compositional data analysis in practice

Organizer and Chair: Valentin Todorov (UNIDO Vienna, AUSTRIA)

Partial least squares for compositional canonical correlation
Michele Gallo (Università di Napoli “L’Orientale”, ITALY)Violetta Simonacci (Università di Napoli “L’Orientale”, ITALY)Massimo Guarino (Università di Napoli “L’Orientale”, ITALY)

Classification with weighted compositions
Karel Hron (Palacký University Olomouc, CZECH REPUBLIC)
Julie Rendlova (Palacký University Olomouc, CZECH REPUBLIC) – Peter Filzmoser (Vienna University of Technology, AUSTRIA)

A compositional analysis approach assessing the spatial distribution of trees in Guadalajara, Mexico
Marco Antonio Cruz, (Universitat Politècnica de Catalunya, SPAIN)
Maribel Isabel Ortego, (Universitat Politècnica de Catalunya, SPAIN) – Elisabet Roca, (Universitat Politècnica de Catalunya, SPAIN)

Room 2

Mathematical theory in clustering

Organizer and Chair: Christian Hennig (Università di Bologna, ITALY)
Discussant: Claudio Agostinelli (Università di Trieste, ITALY)

Asymptotics for bandwidth selection in nonparametric clustering
Alessandro Casa (Università di Padova, ITALY)José E. Chacón (Universidad de Extremadura, SPAIN)Giovanna Menardi (Università di Padova, ITALY)

Asymptotics for bandwidth selection in nonparametric clustering
Christine Keribin (Université Paris-Sud, FRANCE)

Towards axioms for hierarchical clustering of measures
Philipp Thomann (D One, Zuerich, SWITZERLAND)
Ingo Steinwart (University of Stuttgart, GERMANY)

Main Hall

New trends in robust estimation and data analysis

Organizers: Andrea Cerioli (Università di Parma, ITALY)
Agustín Mayo Iscar (Universidad de Valladolid, SPAIN)

Mixtures of multivariate leptokurtic normal distributions
Luca Bagnato, (Università Cattolica del Sacro Cuore, ITALY)Antonio Punzo, (Università di Catania, ITALY)Maria Grazia Zoia, (Università Cattolica del Sacro Cuore, ITALY)

Contamination and manipulation of trade data: the two faces of customs fraud
Domenico Perrotta, (Joint Research Centre of the European Commission, ITALY)
Andrea Cerasa, (Joint Research Centre of the European Commission, ITALY) – Lucio Barabesi, (University of Siena, ITALY) – Mario Menegatti, (University of Parma, ITALY) – Andrea Cerioli, (University of Parma, ITALY)

MacroPCA: an all-in-one PCA method allowing for missing values as well as cellwise and rowwise outliers
Peter Rousseeuw, (KU Leuven, BELGIUM)

Room 3

Preference rankings

Organizer and Chair: Antonio D’Ambrosio (Università di Napoli Federico II, ITALY)

Sampling and learning Mallows and generalized Mallows models under the Cayley distance: short paper
Ekhine Irurozki (Basque Center for Applied Mathematics, SPAIN)Jose A. Lozano (Intelligent Systems Group, UPV/EHU, SPAIN)Borja Calvo (Basque Center for Applied Mathematics, SPAIN)

Item weighted Kemeny distance for preference data
Mariangela Sciandra (Università di Palermo, ITALY)
Simona Buscemi (Università di Palermo, ITALY) – Antonella Plaia (Università di Palermo, ITALY)

An algorithmic approach to item selection in the bayesian Mallows model
Valeria Vitelli (University of Oslo, NORWAY)
Elja Arjas (University of Helsinki, FINLAND) – Arnoldo Frigessi (University of Oslo, NORWAY)


10:05 – 10:25

Coffee break


10:25 – 11:25

Main Hall

Keynote #2 – Statistical challenges in the analysis of complex responses in biomedicine

Sylvia Richardson, University of Cambridge (UK)

To exploit better the structure of the rich sets of characteristics, such as clinical biomarkers, molecular profiles or detailed ontology records, that are currently being collected on large samples of healthy or diseased individuals, statistical models of the variations within and the interplay between different layers of data can be constructed. Generic Bayesian model building strategies and algorithms have been tailored for this purpose. In this talk, I will discuss three areas: implementing joint hierarchical modelling of a large number of responses and a large number of features to discover features associated with many responses ; analysing tree structured ontology data with application for finding the underlying genetic origin of rare diseases; and characterising network structures using fast Bayesian inference in large Gaussian graphical models. Common statistical issues of accounting for model uncertainty, ability to borrow information for retaining power and scalability of Bayesian computations will be highlighted. Modelling strategies and computations will be illustrated on case studies.

11:25 – 12:40

Invited sessions #5

Room 1

Clustering methods for high dimensional data

Organizer and Chair: Cinzia Viroli (Università di Bologna, ITALY)

Noise resistant clustering of high-dimensional gene expression data
Pietro Coretto (Università di Salerno, ITALY)Angela Serra (Faculty of Medicine and Health Technology, FINLAND) Roberto Tagliaferri (Università di Salerno, ITALY)

Recent advances in model-based clustering of high dimensional data
Claire Gormley (University College Dublin, IRELAND)

High-dimensional model-based clustering via random projections
Laura Anderlucci (Università di Bologna, ITALY)
Francesca Fortunato (Università di Bologna, ITALY) – Angela Montanari (Università di Bologna, ITALY)

Main Hall

Dimensionality reduction and model based synthesis of indicators

Organizer and Chair: Maurizio Vichi (Università di Roma “La Sapienza”, ITALY)

Measuring equitable and sustainable well-being in italian regions. A non-aggregative approach
Leonardo Salvatore Alaimo (Università di Roma “La Sapienza”, ITALY)Filomena Maggino (Università di Roma “La Sapienza”, ITALY)

The PARAFAC model in the maximum likelihood approach
Paolo Giordani (Università di Roma “La Sapienza”, ITALY)Roberto Rocci (Università di Roma Tor Vergata, Università di Roma “La Sapienza”, ITALY) – Giuseppe Bove (Università Roma Tre, ITALY)

Dimensionality reduction via hierarchical factorial structure
Carlo Cavicchia (Università di Roma “La Sapienza”, ITALY)Maurizio Vichi (Università di Roma “La Sapienza”, ITALY)Giorgia Zaccaria (Università di Roma “La Sapienza”, ITALY)

Room 2

Robust procedure based on trimming and forward search

Organizers and Chairs: Andrea Cerioli (Università di Parma, ITALY) – Agustín Mayo Iscar (Universidad de Valladolid, SPAIN)

ACE, AVAS and robust data transformations
Anthony C. Atkinson (London School of Economics, UK)

Labour market analysis through transformations and robust multilevel models
Aldo Corbellini (Università di Parma, ITALY)
Marco Magnani (Università di Parma, ITALY) – Gianluca Morelli (Università di Parma, ITALY)

Weighted likelihood estimation of mixtures
Luca Greco (Università di Salerno, ITALY)
Claudio Agostinelli (Università di Trento, ITALY)

Room 3

Unsupervised methods for mixed-type data

Organizer and Chair: Alfonso Iodice d’Enza (Università di Napoli Federico II, ITALY)

Bootstrap inference for missing data reconstruction
Giuseppina Albano (Università di Salerno, ITALY) – Michele La Rocca (Università di Salerno, ITALY) – Maria Lucia Parrella (Università di Salerno, ITALY) – Cira Perna (Università di Salerno, ITALY)

Composite likelihood inference for simultaneous clustering and dimensionality reduction of mixed-type longitudinal data
Roberto Rocci (Università di Roma Tor Vergata, ITALY – Università di Roma “La Sapienza”, ITALY)
Monia Ranalli (Università di Roma Tor Vergata, ITALY)

Influence of outliers on cluster correspondence analysis
Michel van de Velden (Erasmus University of Rotterdam, HOLLAND)
Alfonso Iodice D’Enza (Università di Napoli Federico II, ITALY) – Lisa Schut (Erasmus University of Rotterdam, HOLLAND)


12:40 – 13:40



13:40 – 14:30

Contributed sessions #3

Room 1

Bayesian networks and their applications

A microblog auxiliary part-of-speech tagger based on bayesian networks
Silvia Golia (Università di Brescia, ITALY) – Paola Zola (Università di Brescia, ITALY)

Bayesian networks for the analysis of entrepreneurial microcredit: evidence from Italy
Lorenzo Giammei (Università di Roma “La Sapienza”, ITALY)
Paola Vicard (Università Roma Tre, ITALY)

Uncertainty in statistical matching by BNs
Daniela Marella (Università Roma Tre, ITALY) – Paola Vicard (Università Roma Tre, ITALY) – Vincenzina Vitale (Università di Roma “La Sapienza”, ITALY)

Main Hall

Consumer analysis

Evaluation of the web usability of the University of Cagliari portal: an eye tracking study
Gianpaolo Zammarchi (Università di Cagliari, ITALY) – Francesco Mola (Università di Cagliari, ITALY)

Modelling consumers’ qualitative perceptions of inflation
Marcella Corduas (Università di Napoli Federico II, ITALY)
Rosaria Simone (Università di Napoli Federico II, ITALY) – Domenico Piccolo (Università di Napoli Federico II, ITALY)

Repeated purchasing in the Italian market of sparkling wines
Francesca Bassi (Università di Padova, ITALY) – Fulvia Pennoni (Università di Milano Bicocca, ITALY) – Luca Rossetto (Università di Padova, ITALY)

Room 2

Exclusive and trajectory clustering

Earthquake clustering and centrality measures
Elisa Varini (CNR, ITALY) – Antonella Peresan (Istituto Nazionale di Oceanografia e di Geofisica Sperimentale, ITALY) – Jiancang Zhuang (Institute of Statistical Mathematics, Research Organization of Information and Systems, Tokyo, JAPAN)

Graph-based clustering of visitors’ trajectories at exhibitions
Martina Gentilin (Università di Verona, ITALY)
Pietro Lovato (Università di Verona, ITALY) – Gloria Menegaz (Università di Verona, ITALY) – Marco Cristani (Università di Verona, ITALY) – Marco Minozzo (Università di Verona, ITALY)

Trajectory clustering of moving objects using warping and adaptive euclidean distances
Antonio Irpino (Università della Campania Luigi Vanvitelli, ITALY) – Antonio Balzanella (Università della Campania Luigi Vanvitelli, ITALY)

Room 3

Machine learning and data mining

Machine learning models for forecasting stock trends
Giacomo Camba (Università di Cagliari, ITALY) – Claudio Conversano (Università di Cagliari, ITALY)

Mixtures of experts with flexible concomitant covariate effects: a Bayesian solution
Marco Berrettini (Università di Bologna, ITALY) – Giuliano Galimberti (Università di Bologna, ITALY) – Thomas Brendan Murphy (University College Dublin, IRELAND) – Saverio Ranciati (Università di Bologna, ITALY)

Comparing tree kernels performances in argumentative evidence classification
Davide Liga (Università di Bologna, ITALY)


14:30 – 15:30

Main Hall

Keynote #3 – Unifying Data Units and Models in (Co-)Clustering

Christophe Biernacki, Université Lille 1 (FRANCE)

Statisticians are already aware that any modelling process issue (exploration, prediction) is wholly data unit dependent, to the extend that it should be impossible to provide a statistical outcome without specifying the couple (unit,model). In this talk, this general principle is formalized with a particular focus in model-based clustering and co-clustering in the case of possibly mixed data types (continuous and/or categorical and/or counting features), being also the opportunity to revisit what the related data units are. Such a formalization allows to raise three important spots: (i) the couple (unit,model) is not identifiable so that different interpretations unit/model of the same whole modelling process are always possible; (ii) combining different “classical” units with different “classical” models should be an interesting opportunity for a cheap, wide and meaningful enlarging of the whole modelling process family designed by the couple (unit,model); (iii) if necessary, this couple, up to the non identifiability property, could be selected by any traditional model selection criterion. Some experiments on real data sets illustrate in detail practical beneficits from the previous three spots. It is a joint work with Alexandre Lourme (University of Bordeaux).

15:30 – 15:45

Room 3

Coffee break


15:45 – 17:00

Invited sessions #6

Room 1

Latent variable models for complex networks - II

Organizers: Marco Alfò (Università di Roma “La Sapienza”, ITALY) – Francesco Bartolucci (Università di Perugia, ITALY)
Chair: Daniel Sewell (University of Iowa, US)

Joining factorial methods and blockmodeling for the analysis of affiliation networks
Daniela D’Ambrosio (Università di Napoli Federico II, ITALY)Marco Serino (INVALSI, ITALY)Giancarlo Ragozini (Università di Napoli Federico II, ITALY)

Dynamic clustering of network data: a hybrid maximum likelihood approach
Maria Francesca Marino (Università di Firenze, ITALY)
Maria Silvia Pandolfi (Università di Perugia, ITALY)

A dynamic stochastic block model for longitudinal networks
Catherine Matias (French National Centre for Scientific Research, FRANCE)
Tabea Rebafka (Sorbonne Université, FRANCE) – Fanny Villers (Sorbonne Université, FRANCE)

Main Hall

Methodologies for mixture modeling

Organizer and Chair: Francesca Greselin (Università Milano Bicocca, ITALY)

A stochastic blockmodel for network interaction lengths over continuous time
Michael Fop (University College Dublin, IRELAND)Riccardo Rastelli (University College Dublin, IRELAND)

Mixture modelling with skew-symmetric component distributions
Geoffrey McLachlan (University of Queensland, AUSTRALIA)

A fast and efficient modal EM algorithm for Gaussian mixtures
Luca Scrucca (Università di Perugia, ITALY)

Room 2

Modern likelihood methods

Organizer: Cristiano Varin (Università Ca’ Foscari Venezia, ITALY)
Chair: Michele La Rocca (Università di Salerno, ITALY)

Modern likelihood-frequentist inference at work
Ruggero Bellio (Università di Udine, ITALY)Donald A. Pierce (Oregon State University, US)

Bias reduction for estimating functions and pseudolikelihoods
Nicola Lunardon (Università Milano Bicocca, ITALY)

Computationally efficient inference for latent position network models
Riccardo Rastelli (University College Dublin, IRELAND)
Florian Maire (University of Montreal, CANADA) – Nial Friel (University College Dublin, IRELAND)

Room 3

Statistical challenges in functional data analysis

Organizer and Chair: Tonio di Battista (Università “G. d’Annunzio” Chieti – Pescara, ITALY)

Functional approaches for satellite data clustering and fusion
Claire Miller (University of Glasgow, UK) – Marian Scott (University of Glasgow, UK) – Craig Wilkie (University of Glasgow, UK) – Ruth O’Donnell (University of Glasgow, UK) – Mengyi Gong (British Geological Survey, UK) – Anna Sehn (University of Glasgow, UK)

Assessing social interest in burnout using functional data analysis through google trends
Ana M. Aguilera (Università di Torino, ITALY) – Francesca Fortuna (Università di Torino, ITALY) – Manuel Escabias (Università di Torino, ITALY)

Functional data analysis for spatial aggregated point patterns in seismic science
Elvira Romano (Università della Campania Luigi Vanvitelli, ITALY)
Jonatan A. González (University Jaume I, SPAIN) – Francisco J. Rodríguez Cortés (National University of Colombia, COLOMBIA) – Jorge Mateu (University Jaume I, SPAIN)



Shuttle bus to the Abbey

(guided tour included in the social dinner fees)



Social dinner

Mignano Montelungo Castle



Shuttle bus to Cassino

Day 3 – Friday Sept, 13


8:30 – 9:05

Contributed sessions #4

Room 1

Applications in social science

Foreign immigration and pull factors in Italy: a spatial approach
Oliviero Casacchia (Università di Roma “La Sapienza”, ITALY) – Luisa Natale (Università di Cassino e del Lazio Meridionale, ITALY) – Francesco Giovanni Truglia (ISTAT, ITALY)

Predictive principal components analysis
Simona Balzano (Università di Cassino e del Lazio Meridionale, ITALY) – Maja Bozic (Università di Cassino e del Lazio Meridionale, ITALY) – Laura Marcis (Università di Cassino e del Lazio Meridionale, ITALY) – Renato Salvatore (Università di Cassino e del Lazio Meridionale, ITALY)

Room 2

Clustering of spatial data

Biodiversity spatial clustering
Francesca Fortuna (Università “G. D’Annunzio” di Chieti-Pescara, ITALY) – Fabrizio Maturo (Università della Campania Luigi Vanvitelli, ITALY) – Tonio Di Battista (Università “G. D’Annunzio” di Chieti-Pescara, ITALY)

Post processing of two dimensional road profiles: variogram scheme application and sectioning procedure
Mauro D’Apuzzo (Università di Cassino e del Lazio Meridionale, ITALY) – Rose-Line Spacagna (Università di Cassino e del Lazio Meridionale, ITALY) – Azzurra Evangelisti (Università di Cassino e del Lazio Meridionale, ITALY) – Daniela Santilli (Università di Cassino e del Lazio Meridionale, ITALY)
Vittorio Nicolosi (Università di Roma Tor Vergata, ITALY)

Main Hall

Dimension reduction and visualization

A boxplot for spherical data
Davide Buttarazzi (Università di Cassino e del Lazio Meridionale, ITALY) – Giuseppe Pandolfo (Università di Napoli Federico II, ITALY) – Giovanni C. Porzio (Università di Cassino e del Lazio Meridionale, ITALY) – Christophe Ley (Ghent University, BELGIUM)

Topological stochastic neighbor embedding
Nicoleta Rogovschi (University of Paris 5, FRANCE) – Nistor Grozavu (Université de Paris 13, FRANCE)
Basarab Matei (Université de Paris 13, FRANCE) – Younès Bennani (Université de Paris 13, FRANCE) – Seiichi Ozawa (Kobe University, JAPAN)

Room 3

Partial least squares and dependence structures

A new proposal for building immigrant integration composite indicator
Mario Fordellone (Università di Roma “La Sapienza”, ITALY)

Copula-based non-metric unfolding on augmented data matrix
Marta Nai Ruscone (LIUC, ITALY) – Antonio D’Ambrosio (Università di Napoli Federico II, ITALY)


9:05 – 10:20

Invited sessions #7

Room 1

Directional data analysis

Organizer and Chair: Giuseppe Pandolfo (Università di Napoli Federico II, ITALY)

Local fitting of angular variables observed with error
Marco Di Marzio (Università “G. D’Annunzio” Chieti – Pescara, ITALY)Stefania Fensore (Università “G. D’Annunzio” Chieti – Pescara, ITALY)Agnese Panzera (Università di Firenze, ITALY)Charles C. Taylor (University of Leeds, UK)

Projection-based uniformity tests for directional data
Eduardo García-Portugués (University Carlos III, Madrid, SPAIN)
Paula Navarro-Esteban (University of Cantabria, SPAIN) – Juan Antonio Cuesta-Albertos (University of Cantabria, SPAIN)

Cylindrical hidden Markov fields
Francesco Lagona (Università di Roma Tre, ITALY)

Room 2

Engendering statistics in social studies

Organizer and Chair: Mariangela Zenga (Università Milano Bicocca, ITALY)

The gender parity index for the academic students progress
Adele H. Marshall (Queen’s university of Belfast, UK)
Aglaia Kalamatianou (Panteion University, GREECE)

Exploring gender gap in international mobility flows though a network analysis approach
Ilaria Primerano, (Università di Salerno, ITALY)
Marialuisa Restaino, (Università di Salerno, ITALY)

Evaluating the recruiters’ gender bias in graduate competencies
Paolo Mariani (Università Milano Bicocca, ITALY)Andrea Marletta (Università Pisa, ITALY)

Room 3

Large scale data analysis

Organizer and Chair: Michele La Rocca (Università di Salerno, ITALY)
Discussant: Francesco Palumbo (Università di Napoli Federico II, ITALY)

Recent advancement in neural network analysis of biomedical big data
Pietro Liò (University of Cambridge, UK)Giovanna Maria Dimitri (University of Cambridge, UK)Chiara Sopegno (Politecnico di Torino, ITALY)

Large scale social and multilayer networks
Matteo Magnani (Uppsala University, SWEDEN)

Structure discovering in nonparametric regression by the GRID procedure
Francesco Giordano (Università di Salerno, ITALY) – Soumendra Lahiri (North Carolina State University, US) – Maria Lucia Parrella (Università di Salerno, ITALY)

Main Hall

Mixture modelling for complex data

Organizer and Chair: Michael Fop (University College Dublin, IRELAND)

Gaussian parsimonious clustering models with covariates and a noise component
Keefe Murphy (University College Dublin, IRELAND)Brendan Murphy (University College Dublin, IRELAND)

Modelling unobserved heterogeneity of ranking data with the bayesian mixture of extended Plackett-Luce models
Cristina Mollica (Università di Roma “La Sapienza”, ITALY)
Luca Tardella (Università di Roma “La Sapienza”, ITALY)

Composite likelihood inference for simultaneous clustering and dimensionality reduction of mixed-type longitudinal data
Antonello Maruotti (Università di Roma LUMSA, ITALY)
Monia Ranalli (Università di Roma Tor Vergata, ITALY) – Roberto Rocci (Università di Roma “La Sapienza”, ITALY – Università di Roma Tor Vergata, ITALY)


10:20 – 10:40

Coffee break


10:40 – 10:55

Main Hall

Kick-off data competition (powered by TIM)

Simone Borra, Università di Roma Tor Vergata (ITALY) – Michele Vecchione, TIM (ITALY)


10:55 – 11:55

Main Hall

Keynote #4 – Bayesian Model-Based Clustering with Flexible and Sparse Priors

Bettina Gruen, Johannes Kepler Universitat Linz (AUSTRIA)

Finite mixtures are a standard tool for clustering observations. However,
selecting the suitable number of clusters, identifying cluster-relevant
variables as well as accounting for non-normal shapes of the clusters are
still challenging issues in applications. Within a Bayesian framework we
indicate how suitable prior choices can help to solve these issues. We
achieve this considering mainly prior distributions that have the
characteristics that they are conditionally conjugate or can be
reformulated as hierarchical priors, thus allowing for simple estimation
using MCMC methods with data augmentation.

11:55 – 13:10

Invited sessions #8

Room 1

Data analysis for complex networks

Organizer and Chair: Maria Prosperina Vitale (Università di Salerno, ITALY)
Discussant: Matteo Magnani (Uppsala University, SWEDEN)

Network effect on individual scientific performance: a longitudinal study on the italian scientific community
Domenico De Stefano (Università di Trieste, ITALY)Giuseppe Giordano (Università di Salerno, ITALY)Susanna Zaccarin (Università di Trieste, ITALY)

Clustering of ties defined as symbolic data
Luka Kronegger (University of Lubiana, SLOVENIA)

The structure, evolution and interaction of multiplex networks of scientific collaboration at a research university
Leone Valerio Sciabolazza (Università di Napoli “Parthenope”, ITALY)
Raffaele Vacca (University of Florida, US)Till Krenz (Bureau of Economic and Business Research, Florida, US)

Room 2

Regularization for mixture estimation based on constraints

Organizer and Chair: Agustín Mayo Iscar (Universidad de Valladolid, SPAIN)

Supervised learning in presence of outliers, label noise and unobserved classes
Andrea Cappozzo, (Università Milano Bicocca, ITALY)Francesca Greselin, (Università Milano Bicocca, ITALY)Thomas Brendan Murphy, (University College Dublin, IRELAND)

Penalized vs constrained maximum likelihood approaches for clusterwise linear regression modeling
Roberto di Mari, (Università di Catania, ITALY)
Stefano Antonio Gattone, (Università di Chieti-Pescara “G. D’Annunzio”, ITALY) – Roberto Rocci, (Università di Roma Tor Vergata, ITALY – Università di Roma “La Sapienza”, ITALY)

Robust parsimonious clustering models
Luis Angel Garcia Escudero, (Universidad de Valladolid, SPAIN) – Agustin Mayo Iscar, (Universidad de Valladolid, SPAIN) – Marco Riani, (Università di Parma, ITALY)

Room 3

Statistical analysis of data streams

Organizer and Chair: Antonio Balzanella (Università della Campania Luigi Vanvitelli, ITALY)

Clustering of complex data stream based on barycentric coordinates
Parisa Rastin (Université Paris 13, FRANCE)Basarab Matei (Université Paris 13, FRANCE)Guénaël Cabanes (Université Paris 13, FRANCE)

Unsupervised fuzzy classification for detecting similar functional objects
Fabrizio Maturo (Università della Campania Luigi Vanvitelli, ITALY)
Fortuna Francesca (Università di Chieti e Pescara G. D’Annunzio, ITALY) – Tonio Di Battista (Università di Chieti e Pescara G. D’Annunzio, ITALY)

Co-clustering high dimensional temporal sequences summarized by histograms
Rosanna Verde (Università della Campania Luigi Vanvitelli, ITALY)
Antonio Balzanella (Università della Campania Luigi Vanvitelli, ITALY) – Antonio Irpino (Università della Campania Luigi Vanvitelli, ITALY)

Main Hall

Semiparametric methods for hierarchical data

Organizer and Chair: Anna Maria Paganoni (Politecnico di Milano, ITALY)



Bivariate semi-parametric mixed-effects models for classifying the effects of Italian classes on multiple student achievements
Chiara Masci (Politecnico di Milano, ITALY)Francesca Ieva (Politecnico di Milano, ITALY)Tommaso Agasisti  (Politecnico di Milano, ITALY)Anna Maria Paganoni (Politecnico di Milano, ITALY)

Tree embedded linear mixed models
Anna Gottard (Università di Firenze, ITALY)
Leonardo Grilli (Università di Firenze, ITALY) – Carla Rampichini (Università di Firenze, ITALY) – Giulia Vannucci (Università di Firenze, ITALY)

Use of multi-state models to maximise information in pressure ulcer prevention trials
Linda Sharples (London School of Hygiene and Tropical Medicine, UK)
Isabelle Smith (University of Leeds, UK) – Jane Nixon (University of Leeds, UK)


13:10 – 14:10



14:10 – 15:10

Main Hall

Keynote #5 – Grinding massive information into feasible statistics: current challenges and opportunities for data scientists

Francesco Mola, Università di Cagliari (ITALY)


Massive amounts of data used to make quicker, better and more intelligent decisions to create business value are nowadays available for companies and organizations. Terms like big data, data science, analytics, artificial intelligence, machine learning etc., are very common in both academia and industry. All these areas of research are orientated towards answering the increasing demand for understanding trends and/or discovering patterns in data. Usually, collected data is massive and uncertain due to noise, incompleteness and inconsistency. The main goal of a statistician/data scientist is therefore to turn massive data into feasible information, the latter intended as able to describe efficiently an observed phenomenon, to gain indications about its future evolution as well as to provide useful insights for the ongoing decisional process. All these considerations lead towards arguing that the role of the statistician/data scientist considerably evolved in the latest years.

In my presentation, after a brief description of the scenario summarized above, I will discuss three examples/case studies concerning image validation, hotels’ reputation and social media popularity trying to give a contribution to the debate about turning the enormous amount of available data into feasible statistics. In all cases, ad-hoc but standard classification methods are used to obtain information that is extremely feasible and adds value to a decisional process.


15:10 – 15:20

Main Hall

Closing Statements

Francesca Greselin, Chair Scientific Committee | Giovanni C. Porzio, Chair Local Organizing Committee