Skip to main content

Über dieses Buch

This book is devoted to Corrado Gini, father of the Italian statistical school. It celebrates the 50th anniversary of his death by bearing witness to the continuing extraordinary scientific relevance of his interdisciplinary interests. The book comprises a selection of the papers presented at the conference of the Italian Statistical Society, Statistics and Demography – the Legacy of Corrado Gini, held in Treviso in September 2015. The work covers many topics linked to Gini’s scientific legacy, ranging from the theory of statistical inference to multivariate statistical analysis, demography and sociology. In this volume, readers will find many interesting contributions on entropy measures, permutation procedures for the heterogeneity test, robust estimation of skew-normal parameters, S-weighted estimator, measures of multidimensional performance using Gini’s delta, small-sample confidence intervals for Gini’s gamma index, Bayesian estimation of the Gini-Simpson index, spatial residential patterns of selected foreign groups, minority segregation processes, dynamic time warping to study cruise tourism, and financial stress spill over. This book will appeal to all statisticians, demographers, economists, and sociologists interested in the field.



Entropy Measures: An Health Care Study

In medical emergency situations, the triage process allows patients in potentially life-threatening condition to receive the fastest and most appropriate medical treatment. Triage consists in an evaluation of patients’ medical condition on a colour-based scale, reflecting from major to minor urgency. Shannon’s entropy measures are applied to such process in order to evaluate concordance, overestimation and underestimation of triage codes assigned to patients in two different moments and by different health-care professionals: during the acceptance phase, by nurses (variable X), and by physicians after deepened diagnostic evaluation (variable Y). Entropy indexes were also used to compare the years 2016 and 2015, showing a little increment of equivocal transmission with respect to year 2015.
Enrico Ciavolino, Corrado Crocetta, Amjad D. Al-Nasser

A Review on Heterogeneity Test: Some Permutation Procedures

When dealing with categorical data, generally the notion of heterogeneity may be used instead of that of variability. There are many fields where data may be only represented by nominal categorical variables or by ordinal variables, e.g. opinion polls, performance qualitative assessments, psycho aptitude tests and so on. In this paper we provide a review of some nonparametric methods concerning testing for heterogeneity, based on permutation procedures. Examples of real applications in different frameworks are also shown.
Stefano Bonnini, Eleonora Carrozzo, Luigi Salmaso

Robust Estimation of Skew-Normal Parameters with Application to Outlier Labelling

We suggest to estimate the parameters of the skew-normal distribution by the method of moments, modified so as to achieve robustness. A type of trimmed estimator is used, with the trimming fraction depending on a given scaled deviation from the center. An application to outlier labelling is illustrated.
Mario Romanazzi

Asymptotics of S-Weighted Estimators

The paper studies S-weighted estimator - a combination of S-estimator and the least weighted squares. The estimator allows to adjust the properties, namely the level of robustness of estimator in question to the processed data better than the S-estimator or the least weighted squares can do. The paper offers the proof of its \(\sqrt{n}\)-consistency.
Jan Ámos Víšek

Gini’s Delta to Measure Intensity of Multidimensional Performance

In the present paper we suggest the use of Alkire-Foster [1, 3] dual cut-off method as a measure of multidimensional performance and a modified version of Gini’s delta to measure the concentration of achievements and/or deprivations. On either tails, the information on the joint distribution of achievements or deprivations can be used to measure their concentration. In particular we suggest the use of Gini’s delta to measure concentration within a same tail - for different second order cut-off values k. We call this measure a measure of performance intensity. As an example we apply it to both tails of an OECD regional well-being performance indicator, and to the Multidimensional Poverty Index (MPI).
Silvia Terzi, Luca Moroni

Frequentist and Bayesian Small-Sample Confidence Intervals for Gini’s Gamma Index in a Gaussian Bivariate Copula

In this paper we consider frequentist and Bayesian likelihood-based small-sample procedures to compute confidence intervals for Gini’s gamma index in the bivariate Gaussian copula model. We furthermore discuss how the method straightforwardly extends to any measure of concordance which is available in closed form, and to any type of copula for which the considered measure of concordance has a closed-form expression.
Valentina Mameli, Alessandra R. Brazzale

Bayesian Estimation of Gini-Simpson’s Index Under Mainland-Island Community Structure

The mainland-island community structure is an ecological transposition of a popular model in population genetics in which a fixed number of subpopulations (islands) are connected, through differing immigration rates, to a single metapopulation (mainland) where diversity is generated through speciation. It has been recently shown that a large class of neutral models with this particular structure converges in the large population limit to the Hierarchical Dirichlet process. This finding provides the analogous, in the multipopulation setting, of the Ewens sampling formula for the single population neutral hypothesis. Here we apply some recent results for conditional moments of diversity measures under Gibbs-type priors to derive a Bayesian nonparametric estimator of Gini-Simpson’s index under the Hubbell Unified Neutral Theory of Biodiversity and Biogeography. Potential applications are also illustrated.
Annalisa Cerquetti

Spatial Residential Patterns of Selected Foreign Groups. A Study in Four Italian Cities

What are the spatial residential patterns of the main foreign groups residing in some large Italian cities? Using data from the last Italian demographic census (2011) at sub-municipality level, the study investigates on this research question. A spatial approach is applied to analyze the geographical distribution of the main foreign groups enumerated in the cities of Milan, Rome, Naples and Palermo. The results provide some interesting insights: the distribution of foreign groups coming from central and eastern European countries is quite scattered and shows a comparative low level of dissimilarity to the spatial distribution of Italians. Conversely, foreign groups coming from more distant countries (like China, Bangladesh and Sri Lanka) show spatial distributions characterized by a comparative low level of dispersion and a comparative high level of dissimilarity to the spatial distribution of Italians.
Federico Benassi, Fabio Lipizzi

Minority Segregation Processes in an Urban Context: A Comparison Between Paris and Rome

The process of minority segregation within global cities is a complex phenomenon. In the European urban context, the process of minority segregation seems to differ in the old immigration countries (France, UK, Netherlands, Germany) from the new immigration countries (Italy, Greece, Spain, Portugal). The analysis compares two global cities (Paris and Rome), taking into consideration the current evolution of the minority segregation pattern. The study shows that in both cities no traces of significant increasing segregation emerge over the last twenty years.
Oliviero Casacchia, Luisa Natale, Gregory Verdugo

Similarity of GPS Trajectories Using Dynamic Time Warping: An Application to Cruise Tourism

The aim of this research is to propose an analysis of the trajectories of cruise passengers at their destination using Dynamic Time Warping algorithm. Data collected by means of GPS devices relating to the behavior of cruise passengers in the port of Palermo have been analyzed in order to show similarities and differences among their spatial trajectories at destination. A cluster analysis has been performed in order to identify segments of cruise passengers, based on the similarity of their trajectories. The results have been compared in terms of several metrics derived from GPS tracking data in order to validate the proposed approach. Our findings are of interest from a methodological perspective concerning the analysis of GPS data and the management of cruise tourism destinations.
Mauro Ferrante, Christian Bongiorno, Noam Shoval

The Financial Stress Spillover: Evidence from Selected Asian Countries

The objective of the study is to analyze financial stress spillover among selected Asian countries, namely, China, Pakistan, Sri Lanka, Malaysia and India for the period from Jan 2001 to Dec 2009. The financial stress is measured by Financial Stress Index (FSI), a specially designed comprehensive measure of financial stress. The methodology of Yimlam 2012 is adopted for analyzing dynamics of variance decomposition among countries using FSI for the selected countries. The results of the study confirm that China and Pakistan are the largest transmitters of spillover towards other selected countries. Also the net spillover of China and Pakistan indicated to be positive whereas all other countries show up negative net spillovers. The economic and geographic linkages are suggested to be responsible for influencing magnitude of spillover among selected countries. Finally, the response of each country to shocks in other countries is found to be positive.
Zulfiqar Ali Shah, Muhammad Ejaz Majeed, Biagio Simonetti, Corrado Crocetta
Weitere Informationen

Premium Partner