Seventh Scientific Meeting of the
CLAssification and Data Analysis Group
of the Italian Statistical Society
Università di Catania (Italy) - September 9-11, 2009

Scientific Program

The list of the keynote lectures, Invited Papers, Contributed Papers and Posters to be presto be presented at the Cladag 2009 is the following.


KeyNote Papers


A.  Cerioli

A guided tour of multivariate outlier detection with and without high-breakdown estimators

(Aula Magna, 11/09/09)


J. C. Gower

Extended canonical analysis: theory and applications

(Aula Magna,09/09/09)


D. B. Rubin

Recent issues in causal inference

(Aula Magna,10/09/09)








Specialized Session 1 (09/09/09)

Multivariate Relations for General Data

Organizer: Carles M. Cuadras


C. M. Cuadras
Distance-based approach in multivariate association

K. Gibert, G. Rodríguez-Silva, A. García-Rudolph, J. M. Tormos
On the benefits of mixed AI and Statistics methods for KDD in real-World domains. Clustering based on rules and extensions

J. Graffelman
A universal procedure for biplot calibration





Specialized Session 2 (09/09/09)

Estimating Causal Effects from Observational Data

using Large Databases and Multivariate Matching

Organizer: Fabrizia Mealli


B. Arpino, F. Mealli

The specification of the propensity score in multilevel observational studies


F. Camillo, I. D'Attoma

A multivariate approach to assess balance of categorical covariates in observational studies


S. M. Iacus, G. King, G. Porro

Coarsened exact matching




Specialized Session 3 (09/09/09)

New Developments in Complex Data Modeling for Evaluation

Organizer: Giorgio Vittadini


R. Brüggemann, M. Fattore, J. Owsinski

Using poset theory to compare fuzzy multidimensional poverty across regions


S. C. Minotti

Some notes on the use of cluster-weighted modeling in effectiveness studies


F. Pennoni, F. Bartolucci

Impact evaluation of job training programs by a latent variable model




Specialized Session 4 (09/09/09)

Classification of Measurement Errors: Theory, ISO Standards and Applications

Organizers: Laura Deldossi, Diego Zappa


L. Deldossi, D. Zappa

Measurement errors and uncertainty: a statistical perspective


D. Koltai, R. S. Kenett and D. Kristt

Measurement uncertainty in quantitative chimerism monitoring after stem cell transplantation


S. Salini, N. Solaro

Controlled calibration in presence of clustered measures









Specialized Session 5 (09/09/09)

Visualization of Relationships by Multidimensional Scaling

Organizer: Akinori Okada


G. Bove

Analysis of skew-symmetry in proximity data


M. Nakai

Social stratification and consumption patterns: cultural practices and lifestyles in Japan


A. Okada

Centrality of asymmetric social network: singular value decomposition, conjoint measurement, and asymmetric  multidimensional scaling




Specialized Session 6 (09/09/09)

Computational and Theoretical Aspects in Mixture Models

Organizer: Donatella Vicari


M. Alfò, L. Nieddu, D. Vicari

Regression models for multivariate mixed responses


D. G. Calò, C. Viroli

A finite mixture model for multilevel data

C. Hennig

How to merge Gaussian mixture components




Specialized Session 7 (10/09/09)

Recent Trends in Cluster Analysis

Organizer: Yves Lechevallier


Z. Assaghir, M. Kaytoue, N. Messai, A. Napoli

On the mining of numerical data with Formal Concept Analysis and similarity


M. Chavent, V. Kuentz, J. Saracco

A partitioning method for the clustering of qualitative variables


A. Hardy

Validation of a clustering structure for symbolic data









Specialized Session 8 (10/09/09)

Models for the Analysis of Volatility in Financial Markets

Organizer: Edoardo Otranto



G. Barone-Adesi

Volatility dynamics: escaping the straitjacket


P. Coretto, M. La Rocca, G. Storti

Group structured volatility


G. Giovannetti, M. Velucchi

A MEM analysis of African financial markets




Specialized Session 9 (10/09/09)

Model selection in Advanced Classification and Regression Methods

Organizers: Roberta Siciliano, Claudio Conversano


C. Conversano, E. Dusseldorp

Model selection in classification trunk approach


A. D'Ambrosio, W. J. Heiser

Decision trees for preference rankings


J. S. Rao, J. Jiang and T. Nguyen

A brief overview of fence methods for mixed model selection




Specialized Session 10 (10/09/09)

Operative Problems of Data Mining and Statistical Solutions

Organizer: Furio Camillo


H. Bozdogan, J. A. Howe, S. Katragadda, C. Liberati

Misspecification resistant model selection using information complexity with applications


G. Galimberti, M. Pillati, G. Soffritti

Some issues on robustness of regression trees


G. Menardi, N. Torelli

Some issues in building and assessing classification rules with extremely skewed datasets




Specialized Session 11 (10/09/09)

Statistical Methods for the Assessment of Scientific Research Communities

Organizer: Giuseppe Giordano


C. Davino, F. Palumbo, D. Vistocco

Environmental factors and scientific production: a statistical analysis of MUR data


D. De Stefano, M. P. Vitale

Exploring the pattern of scientific collaboration networks


L. Leydesdorff, Y. Sun

The development of scientific co-authorship networks in national and international contexts




Specialized Session 12 (10/09/09)

Stable Likelihood Methods in Mixture Models

Organizer: Wilfried Seidel


C. Biernacki

Managing degeneracy in Gaussian mixtures


S. Ingrassia, R. Rocci

Constrained EM trajectories for mixtures of normal distributions


K. Sever

Dealing with spurious solutions in finite mixture models




Specialized Session 13 (11/09/09)

Statistical Analysis of Microarray Data

Organizer: Angelo M. Mineo


M. Alfó, F. Martella

Identifying partitions of genes and tissue samples in microarray data


L. Augugliaro, E. C. Wit

Generalizing LARS algorithm using differential geometry


E. C. Wit, L. Augugliaro

Mixed modelling ideas for microarray data




Specialized Session 14 (11/09/09)

Robust Methods for the Analysis of Complex Data

Organizer: Marco Riani


T. DiCiccio, A. C. Monti

Flexible models and robustness


L. Grossi, F. Laurini

Robust portfolio asset  allocation


A. Marazzi

Robust regression with asymmetric errors





Solicited Session 1 (09/09/09)

Environmental Assessment and Air Quality Indicators

Organizer: Antonella Plaia


F. Bruno, D. Cocchi

Interpreting air quality indices as random quantities


D. Lee, C. Ferguson, E. M. Scott

Air quality indicators in health studies


M. Ruggieri, A. Plaia, A. L. Bondì

Aggregated air pollution index: a new proposal




Solicited Session 2 (09/09/09)

Variable Selection in Multivariate Analysis

Organizer: Gabriele Soffritti


N. Dean, A. E. Raftery

Latent class analysis variable selection


C. Maugis, G. Celeux, M. Martin-Magniette

Variable selection in model-based clustering: a general variable role modeling


L. Scrucca

A geometric approach to subset selection and sparse sufficient dimension reduction





Solicited Session 3 (09/09/09)

Statistical Models for Electoral Flow Data

Organizer: Venera Tomaselli


J. M. Bernardo

Forecasting the final results in election day: a Bayesian analysis


A. Forcina, G. M. Marchetti

The Brown and Payne model of voter transition revisited


V. Tomaselli, R. D'Agata

Using multilevel models to analyze the context of electoral data




Solicited Session 4 (10/09/09)

Partitioning Methods for Statistical Learning

Organizer:  Massimo Aria


G. Boari, G. Cantaluppi

Further considerations on latent segmentation techniques for customer heterogeneity detection


G. Giordano, R. Remmerswaal

Non-parametric regression model for a hierarchical data-structure: a comparison with the classical approaches


G. Soffritti, G. Galimberti

Detecting multiple cluster structures through model-based clustering methods




Solicited Session 5 (10/09/09)

Identification and Estimation of Causal Effects in the Presence of Complications

Organizer: Leonardo Grilli


P. Frumento, B. Pacini

Causal inference with multivariate outcome:  a simulation study


A. Mattei

Identifying and estimating direct and indirect effects


G. Roli, B. Pacini,

Evaluating the effects of financial aids to firms with nonignorably missing outcomes





Solicited Session 6 (10/09/09)

Statistical Analysis of Social Networks of Enterprises

Organizer:  Francesco Palumbo


P. Ameli, F. Niccolini, F. Palumbo

Latent ties identification in inter-firms social networks


M. R. D’Esposito, A. Del Monte, G. Giordano, M. P. Vitale

The analysis of structural characteristics of innovative networks


S. Liverani, A. Petrucci

Spatial clustering of multivariate data using weighted MAX-SAT




Solicited Session 7 (10/09/09)

Methodology for Evaluating the Effectiveness of the Italian University System

Organizer:  Luigi Fabbris


M. Bini

Evaluating the External Effectiveness of the University Education in Italy


G. Elias

How to approach the evaluation of the effectiveness of higher education in Italy


L. Fabbris, G. Boccuzzo, M. C. Martini, M. Scioni

A participative process for the definition of a human capital indicator




Solicited Session 8 (10/09/09)

Advances in Data Collection for Surveys

Organizer:  Pier Francesco Perri


G. Diana

Calibration estimators in randomized response models


I. Krumpal

Estimating the prevalence of unsocial opinions via the Randomized Response Technique


P. G. M. van der Heijden, U. Böckenholt

Randomized response, statistical disclosure control and e-commerce




Solicited Session 9 (10/09/09)

Functional Data Analysis

Organizer:  Rosanna Verde


G. Costanzo, F. Dell'Accio, G. Trombetta

Prediction of an industrial kneading process via the adjustment curve


 T. Di Battista, S. A. Gattone, A. De Sanctis

Dealing with FDA estimation methods


A. Goia

Functional linear regression: applications to time series prediction




Solicited Session 10 (11/09/09)

Econometric Analysis of Data on Market Behaviors

Organizer: Roberto Cellini



V. Adorno, C. Bernini, G. Pellegrini

Productivity, efficiency and public subsidies. An empirical evaluation


L. Arciero, A. Mercatanti

Evaluation of the effect of debit cards on cash demand by propensity score regressions in a principal stratification model


R. Cellini, L. Lambertini, A. Sterlacchini

Managerial incentive and the firms’ propensity to invest in product and process innovation


M. Di Giacomo, M. Piacenza, G. Turati

The effectiveness of “flexible” specific taxation in stabilizing fuel prices. An empirical investigation on the Italian wholesale markets




Solicited Session 11 (11/09/09)

From Market Segmentation to Different Customer Loyalties

Organizer:  Mario Montinaro


P. Chirico, A. Lo Presti

Customer loyalty analysis in an heterogeneous market: a comparison between a priori

segmentation and model-based segmentation


P. Ferrari, A. Barbiero , G. Manzi

Missing data handling in  setting up indicators from categorical data


M. Montinaro, I. Sciascia

Probabilistic models for market segmentation and customer loyalty


G. Nicolini, L. Dalla Valle

The problem of non random samples in customer satisfaction surveys




Solicited Session 12 (11/09/09)

Markov and Hidden Markov Processes for Data Analysis and Simulation

Organizer:  Isabella Morlini



F. Bartolucci, A. Farcomeni, F. Pennoni

Analysis of longitudinal data via latent Markov models and its extensions


M. Costa, L. De Angelis

A dynamic analysis of stock markets through latent Markov models


A.  Maruotti, R. Rocci

A semiparametric approach to mixed nonhomogeneous hidden Markov models


R. Paroli

Hidden Markov processes for image data analysis








Contributed Session 1 (09/09/09)

Methods and Applications in Cluster Analysis



F. Campobasso, A. Fanizzi, M. Tarantini

A proposal for a fuzzy approach to clustering techniques


R. Guido, M. Misuraca, F. Vocaturo

Document classification in public administration: an automatic strategy for digital protocol


A. Lourme, C. Biernacki

Joint clustering of two samples from multiple origins by equalizing their entropy


S. C. Minotti, G. Spedicato

A statistical R package for Cluster-Weighted Modeling


A. Pastore, S. Tonellato

A class of indexes for comparing model-based clustering solutions





Contributed Session 2 (09/09/09)

Data Analysis for the Assessment of Public Administrations


M. Finocchiaro Castro, C. Guccio
Efficiency analysis of heritage authorities: a DEA-bootstrap approach

M. Gnaldi, M. Ranalli

Ranking the Italian universities: a sensitivity analysis of research performance composite indicators


P. Mariani, E. Zavarrone

The measure of economic re-evaluation: a coefficient based on conjoint analysis


F. Mealli, C. Rampichini

Exploiting nonlinear difference-in-difference assumptions in a regression discontinuity design


M. Porcu, I. Sulis, N. Tedesco

Evaluating lecturer’s capability over time. Some evidence from surveys on university course quality




Contributed Session 3 (09/09/09)

Data Analysis in Economics and Finance



G. Amisano, R. Savona

Taming Financial Inaccessibility with Bayesian state space models


C. Cutugno

Model based clustering for solvency analysis


A. Geyer-Schulz, B. Hoser

Financial accounting and eigensystem analysis


Y. Henchiri, S. Benammou, T. Hamza

Using support vector machine approach for forecasting the failure of the Tunisian companie


A. A. Romano, G. Scandurra

Impact of exogenous shocks on oil product market prices




Contributed Session 4 (09/09/09)

Methods and Applications in Regression


G. Adelfio, M. Chiodi

Probabilistic forecast for Northern New Zealand seismic process: a kernel-based approach


F. Crippa, M. Marelli

Mixed-effects modeling for items and subjects in psycholinguistics: an application


A. Mazza, A. Punzo

Graduation of demographic data using discrete beta kernel estimates


I. Mdimagh, S. Benammou

Using partial least squares regression in lifetime analysis


E. Rocco

Kernel-Type Smoothing Methods in the Nonresponse Context




Contributed Session 5 (09/09/09)

Classification and Visualization of Time series


A. Balzanella, R. Verde, Y. Lechevallier

A new approach for clustering multiple streams of data


C. Drago, G. Scepi

Univariate and multivariate tools for visualizing financial time series


A. Tarsitano

A weighted distance for short time series




Contributed Session 6 (09/09/09)

Multivariate Methods and Applications



G. D'Epifanio

A policy-planning based data-analysis in social agents’s worthiness scaling


M. Marozzi

Some notes on indicator variable reduction


V. A. Tutore, A. D'Ambrosio

Three-way data analysis by tree-based partitioning




Contributed Session 7 (09/09/09)

Testing Statistical Models



S. Facchinetti, S. A. Osmetti

The Kolmogorov-Smirnov goodness of fit test for discrete extreme value distributions


F. Greselin, S. Ingrassia, A. Punzo

Multivariate tests for patterned covariance matrices


F. Torti, D. Perrotta

The test size of the forward search for regression outliers




Contributed Session 8 (10/09/09)

Computer Intensive Methods in Data Analysis



A. Brogini, D. Slanzi

Confident Bayesian networks: a non-parametric bootstrap approach


A. Cossari

Using the bootstrap in the analysis of fractionated designs


P. Frumento

Using genetic algorithms in estimation of finite mixture models


M. Vezzoli, P. Zuccolotto

Variable importance measurement within hierarchical data




Contributed Session 9 (10/09/09)

Methods and Applications in Item Response Theory



L. Bertoli-Barsotti, A. Punzo

Modelling missingness in the Rasch paradigm


S. Golia

The impact of uniform and nonuniform differential item functioning on Rasch measure


M. Matteucci, S. Mignani, R. Ricci

Latent variable models to evaluate the final exam in the Italian lower secondary school


M. Matteucci, B. Veldkamp

The inclusion of prior information in CAT administration




Contributed Session 10 (10/09/09)

Statistical  Models in Medicine



S. Arima, V. Pecora, L. Tardella

Identifying epitopes in allergic reactions through hierarchical Bayesian modeling


P. Brutti, S. Gubbiotti, V. Sambucini

A Bayesian two-stage design for phase II clinical trials with both efficacy and safety endpoint


R. Gargano, A. Alibrandi

Modeling resistin levels data by mixture of regression


I. Poli, D. Slanzi, L. Villanova, D. De Lucrezia, G. Minervini, F. Polticelli

Non natural protein structure identification




Contributed Session 11 (11/09/09)

Clustering Functional Data



S. Bande, S. Ghigo, R. Ignaccolo

Classification of municipalities with respect to air quality assessment


F. Bruno, D. Cocchi, F. Greco

Functional cluster analysis for compositional data


E. Romano, A. Balzanella, R. Verde

A clusterwise regression strategy for spatio-functional data



E. Romano, V. Cozza

Clustering spatio-functional data: a method based on a nonparametric variogram estimation for functional data




Contributed Session 12 (11/09/09)

 Copulas and Association Measures



G. Bianchi, F. M. L. Di Lascio, S. Giannerini, A. Manzari, A. Reale, G. Ruocco

Exploring copulas for the imputation of missing nonlinearly dependent data


F. Domma, S. Giordano

Dependence in stress-strength models with fgm copula and Burr III margins


D. Marasini, P. Quatto

Control sample and association


I. Morlini, S. Zani

New weighted similarity indexes




Contributed Session 13 (11/09/09)

Data Analysis for Customer Satisfaction

M. Gallo, D. Fiorenzo
A strategy to evaluate the satisfaction of different passenger group

P. Mariani, E. Zavarrone

Satisfaction, loyalty and WOM in dental care sector


A. M. Oliveri, A. M. Parroco

CAPI versus PAPI and the production of some non-sampling errors


G. Schoier

On the problem of variable reduction in a customer satisfaction context




Contributed Session 14 (11/09/09)

Analysis of Categorical Data



M. Carpita, M. Manisera

On the nonlinearity of homogeneous ordinal variables


M. Grassia, D. Nappo

An algorithm for the treatment of qualitative variables in a SEM model and its validation


B. Simonetti, L. D'Ambra, P. Amenta

New developments in ordinal non symmetrical correspondence analysis


A. Zarraga, B. Goitisolo

Correspondence analysis of surveys with multiple response questions




Contributed Session 15 (11/09/09)

Classification of Spatial Data



A. Altavilla, A. Mazza, M. Mucciardi

Differences between the settlement patterns of individual immigrants in the city of Catania


A. Altavilla, M. Mondello

Mortality in environmental high risk area


G. Aretusi, L. Fontanella, L. Ippoliti, A. Merla

Supervised classification of thermal high-resolution IR images for the diagnosis of Raynaud’s phenomenon


M. Tsuji, S. Konsha, T. Shimokawa

A presentational approach to clustering the regional structural analysis with time




Contributed Session 16 (11/09/09)

Multivariate Analysis


C. E. Bonafede, P. Cerchiello, P. Giudici

A semantic based Dirichlet compound multinomial model


P. Cozzucoli

A control chart to monitor the overall defectiveness of the proces


S. Figini

Local statistical models for variables selection


P. Giudici, E. Raffinetti

On dependence measures in a multivariate context







F. Bassi

Latent class analysis for marketing scales development


D. Lee, T. Neocleous

Using quantile regression in environmental epidemiology


G. Marcon, G. Adelfio, M. Chiodi  

Gamma kernel intensity estimation in space-time point processes

E. Moreno, J. Girón, L. Martínez

Generalizing the BIC: an objective Bayesian approach


S. Pandolfi, F. Bartolucci, N. Friel

Reversible Jump Acceleration via Multiple-try