Spanning trees and identifiability of a single-factor model

The aim of this paper is to propose conditions for exploring the class of identifiable Gaussian models with one latent variable. In particular, we focus attention on the topological structure of the complementary graph of the residuals. These conditions are mainly based on the presence of odd cycles and bridge edges in the complementary graph. We propose to use the spanning tree representation of the graph and the associated matrix of fundamental cycles. In this way it is possible to obtain an algorithm able to establish in advance whether modifying the graph corresponding to an identifiable model, the resulting graph still denotes identifiability.     

Assessing corporate governance quality: substance over form

After the 2008 global financial crisis and corporate scandals, assessing and improving corporate governance quality (CGQ) is essential. This paper proposes a different approach to evaluate CGQ, to overcome the conceptual and methodological limits of the previous rating systems. It tries to go beyond the objectives of the existing models by suggesting an alternative operating model, (aligned with the new CG guidelines) that provides a concise index for monitoring and decision-making. Using a Fuzzy Expert System (FES), the authors propose a formalized model that: (1) represents all the factors (structural and behavioral) that affect the quality of corporate governance in terms of practical and objective decision-making procedure; (2) is a flexible and useful management tool for supporting the “Board review” and assessing the increase in CGQ associated with particular decisions; (3) supervisors can use to assess CG adequacy by replacing or integrating the experts’ opinions with interviews/questionnaires filled in by directors and managers or through direct observation, as recently suggested by EBA/ESMA. This paper highlights the importance of behavioral features and group dynamics in corporate governance and represents them in an integrated model together with other structural and organizational elements.     

Cigarette Smoking and Multiple Health Risk Behaviors: A Latent Class Regression Model to Identify a Profile of Young Adolescents

Cigarette smoking is often established during adolescence when other health‐related risk behaviors tend to occur. The aim of the study was to further investigate the hypothesis that risky health behaviors tend to cluster together and to identify distinctive profiles of young adolescents based on their smoking habits. To explore the idea that smoking behavior can predict membership in a specific risk profile of adolescents, with heavy smokers being more likely to exhibit other risk behaviors, we reanalyzed the data from the 2014 Health Behaviour in School‐Aged Children Italian survey of about 60,000 first‐ and third‐grade junior high school (JHS) and second‐grade high school (HS) students. A Bayesian approach was adopted for selecting the manifest variables associated with smoking; a latent class regression model was employed to identify smoking behaviors among adolescents. Finally, a health‐related risk pattern associated with different types of smoking behaviors was found. Heavy smokers engaged in higher alcohol use and abuse and experienced school failure more often than their peers. Frequent smokers reported below‐average academic achievement and self‐rated their health as fair/poor more frequently than nonsmokers. Lifetime cannabis use and early sexual intercourse were more frequent among heavy smokers. Our findings provide elements for constructing a profile of frequent adolescent smokers and for identifying behavioral risk patterns during the transition from JHS to HS. This may provide an additional opportunity to devise interventions that could be more effective to improve smoking cessation among occasional smokers and to adequately address other risk behaviors among frequent smokers.     

On the estimation of the Lorenz curve under complex sampling designs

Statistical Methods & Applications - This paper focuses on the estimation of the concentration curve of a finite population, when data are collected according to a complex sampling design with...     

Italian Urban Plans: Diversified Approaches and Methods for Assessing Their Quality

This article poses the difficult task of providing a brief overview of Italian urban planning, difficult to understand for foreign observers because each Region has different legislation and urban planning tools. The paper describes an original methodology that evaluates the quality of the approach differentiated as (1) ‘of the plan’ (reformist, participative, from the territory); (2) ‘in the plan’ (integrated strategic, sustainable); (3) ‘with the plan’ (conformative, methodological, recompositional, systemic-infrastructural); and (4) ‘beyond the plan’ (beyond governance methodologies). The paper also compares aspects of quality present in plans for large, medium and small Italian cities. The conclusions review factors of excellence that may be used in other international scenarios.     

Psychometric Properties of a Portuguese Version of the Subjective Happiness Scale

The Subjective Happiness Scale (SHS) constitutes an instrument for assessing subjective happiness. This study aims to present the validation of the SHS in a Portuguese adult population. A large representative sample (1,017 participants), from five different age groups was considered. Configurational invariance of the unidimensional structure of the scale was tested through multi-group confirmatory factorial analysis, and reliability and criterion-validity were investigated. The results support the single factorial structure and reliability of the SHS. Furthermore, the SHS correlated positively with the Satisfaction with Life Scale, thus supporting its criterion validity. Therefore, the Portuguese version of the SHS is a valid instrument to use with an adult Portuguese Sample.     

The management of participatory cultural initiatives: learning from the discourse on intellectual capital

Based on previous research that highlights some gaps in the study of the new phenomena of Participatory Cultural Initiatives (PCIs), this paper addresses the need to understand how the relevance of those initiatives is perceived, tentatively measured and communicated to strategic stakeholders. With this, we want to investigate a possible contribution of the discourse on Intellectual Capital (IC) and Intangible Assets for the cultural sector in general and PCIs in detail. The IC perspective is adopted as a boundary object to analyse the key actors’ behaviours concerning PCIs, following a performative approach. In doing so, this field research applies a case study method of analysis by investigating two PCIs in Italy: Big City Life, a street art district in Tor Marancia (Rome) and the project of Matera as European Capital of Culture 2019. Data have been gathered through both primary (interviews) and secondary (official documents, social media, websites and press releases) sources of evidence. The analysis is mainly developed at a micro (single organisation) and meso (district/city) level. Given the originality of the topic whose research is still at an infancy stage and without a well-grounded theory, we seek to gain significant insights into this new issue by applying an exploratory approach. Through the IC lens, it is possible to gain a better understanding of how the PCIs’ impact is assessed and communicated to its stakeholders. Evidence shows that narratives and visualisations play a crucial role by supporting comprehension of the PCIs’ key aspects and their interwoven relations. A condition for narratives and visualisations to be accepted as legitimation tools is the development of a trust relationship among stakeholders. As a result, the impact of PCIs has dimensions and interwoven aspects which are not amenable for traditional performance measures. In summary, it is possible to state that PCIs trigger a focus on talking and showing to understand, instead of a focus on measuring to understand. The paper discusses how the understanding and management of PCIs can improve with lessons learnt from the discourse on IC and with reference to Habermas’ theory of communicative action. Primarily, we focus on the use of narrative and visual representations by key actors to communicate with strategic stakeholders to gain trust and legitimation.

     

Development and implementation of an operating room scheduling tool: an action research study

In this study, we report the results of an action research project whose aim was to develop and implement an operating room scheduler at the Meyer University Children’s Hospital. The study offers insights into the features that make an MSS optimisation model and scheduler effective and easy to implement, and shed light on those actions facilitating their introduction and use. Specifically, it suggests that creating an effective operating room scheduler requires clustering patients in homogeneous surgery groups and developing a flexible tool that allows: scheduling surgery groups instead of actual patients, easily adding/removing constraints, changing the objective function(s) and adjusting the planning horizon. In addition, it posits that gaining the commitment of top management by showing credible preliminary results, inferring stakeholder preferences by letting them comment on tentative schedules, introducing changes gradually and involving staff at lower levels of the hospital hierarchy can significantly facilitate the scheduler development and implementation.     

Estimation of contingency tables in complex survey sampling using probabilistic expert systems

In this paper we explore the possibility to use a particular class of models, known as probabilistic expert systems, to define two classes of estimators of a contingency table in case of stratified sampling designs. The two classes are characterized by the different role of the sampling design: in the first, the sampling design is treated as an additional variable; in the second, it is used only for estimation purposes by means of the survey weights. The bias/variance trade off of these estimators is analyzed and the consequences of model misspecification are illustrated. Furthermore, it is shown that the Horvitz–Thompson estimator belongs to both classes of estimators. It comes out that the Horvitz–Thompson estimator is almost always inefficient but robust. Monte Carlo simulations illustrate the efficiency of the proposed estimators.     

Analysis and correction of bias in Total Decrease in Node Impurity measures for tree-based algorithms

Variable selection is one of the main problems faced by data mining and machine learning techniques. These techniques are often, more or less explicitly, based on some measure of variable importance. This paper considers Total Decrease in Node Impurity (TDNI) measures, a popular class of variable importance measures defined in the field of decision trees and tree-based ensemble methods, like Random Forests and Gradient Boosting Machines. In spite of their wide use, some measures of this class are known to be biased and some correction strategies have been proposed. The aim of this paper is twofold. Firstly, to investigate the source and the characteristics of bias in TDNI measures using the notions of informative and uninformative splits. Secondly, a bias-correction algorithm, recently proposed for the Gini measure in the context of classification, is extended to the entire class of TDNI measures and its performance is investigated in the regression framework using simulated and real data.