Studies in Theoretical and Applied Statistics: SIS 2021, Pisa, Italy, June 21–25

Preface
Contents
A Composite Index of Economic Well-Being for the European Union Countries
	1 Introduction
	2 Theoretical Framework
	3 Methodological Aspects
	4 A Longitudinal Analysis
	5 Conclusions
	References
A Dynamic Power Prior for Bayesian Non-inferiority Trials
	1 Introduction
	2 Methodology
		2.1 Priors Construction
	3 Application
	4 Frequentist Type-I Error and Power
	5 Sample Size Determination
	6 Conclusions
	References
A Graphical Approach for the Selection of the Number of Clusters in the Spectral Clustering Algorithm
	1 Introduction
	2 Spectral Clustering
	3 A Multi-graphical Approach to Select the Number of Clusters
	4 Numerical Studies
		4.1 Simulation Study: Skew-t Data
		4.2 Synthetic Data: Circles&Squares Dataset
		4.3 Real Data: Anuran Calls Dataset
	5 Conclusions
	References
A Higher-Order PLS-SEM Approach to Evaluate Football Players' Performance
	1 Introduction
	2 The PLS-SEM and the Higher-Order Model
	3 Data and Models Specification
	4 Results and Discussion
	5 Conclusion
	References
A Latent Markov Approach for Clustering Contracting Authorities over Time Using Public Procurement Red Flags
	1 Introduction
	2 Data and Statistical Model
		2.1 BDNCP and Red Flag Indicators in PP
		2.2 The Latent Markov Model
	3 Results
	4 Conclusions
	References
A Multiplex Network Approach for Analyzing University Students' Mobility Flows
	1 Introduction
	2 Multiplex Network Definition and Analysis
	3 Data Description and Normalization Procedure
	4 Main Results
	5 Concluding Remarks
	References
A Statistical Model for Predicting Child Language Acquisition: Unfolding Qualitative Grammatical Development by Using Logistic Regression Model
	1 Introduction
	2 Sampling and Data Optimization Strategy
	3 Statistical Model and Variables Selection
	4 Conclusion and Future Directions
	References
Adaptive COVID-19 Screening of a Subpopulation
	1 Introduction
	2 Methods
		2.1 Modeling Time Evolving Proportions
		2.2 Forecasting Using ARMA Models
		2.3 Detecting Excessive Presence of the Characteristic of Interest in the Subpopulation
		2.4 A Naive Fixed Threshold
	3 A Case Study: COVID-19 Testing at Politecnico di Torino
		3.1 Current Testing
		3.2 A Proposal for Adaptive Testing
		3.3 Operating Characteristics of Adaptive Testing
	4 Discussion
	References
Alternative Probability Weighting Functions in Behavioral Portfolio Selection
	1 Introduction
	2 Prospect Theory
	3 The Probability Weighting Function
	4 A Behavioral CPT Portfolio Selection
	5 A Particle Swarm Optimization Solution Approach
	6 An Application to the European Stock Market
	7 Conclusions
	References
Bayesian Quantile Estimation in Deconvolution
	1 Introduction
	2 Main Results
		2.1 Posterior Convergence Rates for Quantiles in Deconvolution
		2.2 Inversion Inequality for the Kolmogorov Metric
		2.3 Applications
	3 Final Remarks
	References
Can the Compositional Nature of Compositional Data Be Ignored by Using Deep Learning Approaches?
	1 Introduction
		1.1 Compositional Data
		1.2 Expressing Compositional Data in Log-Ratio Coordinates
		1.3 Requirements for Compositional Data Analysis and Difficulties in Practice
	2 Problems and Difficulties in Practice
	3 Artificial Neural Networks
	4 Data Sets, Their Treatment and Selected Methods
	5 Results
	6 Discussion and Conclusion
	References
Citizen Data and Citizen Science: A Challenge for Official Statistics
	1 Introduction: Next Generation Data
	2 The Evolution Process for Producing Statistics and Indicators
	3 Citizen Data and Citizen Science: A Challenge for OS
	4 A Project on the Measure of the Quality of Citizen Data for the Compilation of SDGs Indicators
	References
Detecting States of Distress in Financial Markets: The Case of the Italian Sovereign Debt
	1 Introduction
	2 Empirical Analysis
	3 Conclusions
	References
Forecasting Combination of Hierarchical Time Series: A Novel Method with an Application to CoVid-19
	1 Introduction
	2 Literature Review
	3 The Framework
		3.1 Hierarchical Cross-Sectional Reconciliation: The Chosen Method
		3.2 The Forecast Combination Methods Adopted
		3.3 The Entertained Statistical Models
	4 The Proposed Method
		4.1 The Bias Correction Procedure (Step 8)
	5 Justification of the Method
	6 Empirical Study
		6.1 Performances of the Method
	7 Conclusions and the Future Directions
	8 Data Availability
	9 Disclaimer
	References
Frequency Domain Clustering: An Application to Time Series with Time-Varying Parameters
	1 Introduction
	2 The Clustering Procedure
		2.1 The GAS Model with Gaussian Density
		2.2 Frequency Domain Clustering
	3 Simulation Study
	4 Application to Financial Time Series
	5 Final Remarks
	References
Heterogeneous Income Dynamics: Unemployment Consequences in Germany and the US
	1 Introduction
	2 Job Loss and the Strategies to Buffer Its Consequences
		2.1 Individual Counter-Mobility Strategy
		2.2 Household Buffer
		2.3 Welfare State Buffer
	3 Data and Methods
		3.1 Definition of Measures and Variables
		3.2 The Model—The ‘Distributed’ Fixed-Effect
	4 Results
		4.1 Germany
		4.2 United States
	5 Conclusions
	Appendix
	References
How Much Do Knowledge About and Attitude Toward Mobile Phone Use Affect Behavior While Driving? An Empirical Study Using a Structural Equation Model
	1 Introduction
	2 Literature Review
		2.1 Mobile Phone Behavior
		2.2 Knowledge, Attitudes and Behaviors
	3 Material and Methods
		3.1 Participants and Procedure
		3.2 Statistical Analysis
	4 Results
	5 Discussion
		5.1 Limitations
	6 Conclusions
	References
Impact of Emergency Online Classes on Students’ Motivation and Engagement in University During the Covid-19 Pandemic: A Study Case
	1 Introduction
	2 Data Collection
	3 Results of the Multivariate Analyses
	4 A Study of the Gender Differences and Inequalities
	5 Conclusions
	References
Local Heterogeneities in Population Growth and Decline. A Spatial Analysis of Italian Municipalities
	1 Introduction and Brief Literature Review
	2 The Italian Case
	3 Data and Methods
	4 Global and Local Spatial Autocorrelation of Population Growth and Decline
	5 Results
	6 Conclusions and Further Developments
	References
Model Predictivity Assessment: Incremental Test-Set Selection and Accuracy Evaluation
	1 Introduction
	2 Predictivity Assessment Criteria for an ML Model
		2.1 The Predictivity Coefficient
		2.2 Weighting the Test Sample
	3 Test-set Construction
		3.1 Fully-Sequential Space-Filling Design
		3.2 Support Points
		3.3 Kernel Herding
		3.4 Numerical Illustration
	4 Numerical Results I: Construction of a Training Set and a Test Set
		4.1 Test-cases
		4.2 Results and Analysis
	5 Numerical Results II: Splitting a Dataset into a Training Set and a Test Set
		5.1 Industrial Test-Case CATHARE
		5.2 Benchmark Results and Analysis
	6 Conclusion
	References
Multiple Measures Realized GARCH Models
	1 Introduction
	2 Realized GARCH Models
	3 Adaptive Frequency Realized GARCH Models
	4 Principal Component Realized GARCH Models
	5 Estimation
	6 Data Description
	7 Tail-risk Forecasting
		7.1 Forecasting Design
		7.2 Evaluation of Tail-Risk Forecasts: Backtesting and Scoring Rules
		7.3 Empirical Results
	8 Conclusion
	References
Multiversal Methods in Observational Studies: The Case of COVID-19
	1 Introduction
	2 Theory: Multiversal Methods
		2.1 What Is a Specification?
		2.2 Janus Effect
	3 Materials
		3.1 Why Coronavirus
		3.2 Selection of Covariates
	4 Methods: Multiverse Models
		4.1 Measuring the Outcome of a Shift in alphaα
	5 Results
	6 Discussion
	References
Neural Network for the Statistical Process Control of HVAC Systems in Passenger Rail Vehicles
	1 Introduction
	2 The Proposed Approach
		2.1 Artificial Neural Networks
		2.2 Autoencoder
		2.3 Construction and Control Limit Estimation of the Monitoring Statistics
	3 Real-Case Study
		3.1 HVAC and Data Description
		3.2 Results
	4 Conclusions
	References
On the Use of the Matrix-Variate Tail-Inflated Normal Distribution for Parsimonious Mixture Modeling
	1 Introduction
	2 Methodology
		2.1 The Matrix-Variate Tail-Inflated Normal Distribution
		2.2 Parsimonious Mixtures of Matrix-Variate Tail-Inflated Normal Distributions
		2.3 Parameter Estimation
		2.4 A Note on the Initialization Strategy
	3 Numerical Analyses
		3.1 Simulated Data
		3.2 Real Data
	4 Conclusions
	References
Population Size Estimation by Repeated Identifications of Units. A Bayesian Semi-parametric Mixture Model Approach
	1 Introduction
	2 The Model
		2.1 MCMC Algorithm
	3 Sparse Finite Mixtures
	4 Results on Smuggling Data
	5 Concluding Remarks
	References
Spatial Interdependence of Mycorrhizal Nuclear Size in Confocal Microscopy
	1 Introduction
	2 Materials and Methods
		2.1 Sample Preparation and Confocal Microscopy
		2.2 Image Processing Protocol
	3 Results
		3.1 Density and Nuclei Size
		3.2 CE Index
		3.3 Estimated Variograms
		3.4 Estimation with Kriging
	4 Discussion
	References
Spatially Balanced Indirect Sampling to Estimate the Coverage of the Agricultural Census
	1 Introduction
	2 Theoretical Framework
	3 Spatially Balanced Sampling Design
	4 Proposals and Simulations
	5 Variance Estimation
	6 Conclusion
	References
The Assessment of Environmental and Income Inequalities
	1 Introduction
	2 Methodology
		2.1 Unidimensional Indicators
		2.2 Bidimensional Indicators
	3 A Case Study on Italian Data
		3.1 Unidimensional Indicators
		3.2 Bidimensional Indicators
	4 Conclusions
	References
The Italian Social Mood on Economy Index During the Covid-19 Crisis
	1 Introduction
	2 Background
	3 Data and Methods
		3.1 Social Mood on Economy Index
		3.2 SMEI Derived Series
		3.3 Other Nontraditional High-Frequency Series
	4 Main Results
		4.1 SMEI and SME Covid Share Series
		4.2 The Covid SMEI
	5 Conclusion
	References
The Rating of Journals and the Research Outcomes in Statistical Sciences in Italian Universities
	1 Preliminaries
	2 The Output of Research of the Italian Academic Statistical Community in the Last Decade
	3 Conclusions
	Appendix
	References
Trusted Smart Surveys: Architectural and Methodological Challenges Related to New Data Sources
	1 Introduction
	2 Methodological Aspects of a Trusted Smart Survey
		2.1 Representation and Measurement Errors
		2.2 Smart Data Quality and Management of Error Sources
	3 Modelling an Architectural Framework for TSSu
		3.1 TSSu Business Layer Modelling
		3.2 TSSu Application Layer Modelling
		3.3 TSSu Operational Models
		3.4 Impact of Trust on Platform Requirements
	4 Conclusions
	References
Web Surveys: Profiles of Respondents to the Italian Population Census
	1 Introduction
	2 The Italian Permanent Population and Housing Census
	3 Data
	4 Methods
	5 Results
	6 Conclusions
	References