The power of (extended) monitoring in robust clustering

We complement the work of Cerioli, Riani, Atkinson and Corbellini by discussing monitoring in the context of robust clustering. This implies extending the approach to clustering, and possibly monitoring more than one parameter simultaneously. The cases of trimming and snipping are discussed separately, and special attention is given to recently proposed methods like double clustering, reweighting in robust clustering, and fuzzy regression clustering.     

Impact measurement and dimension reduction of auxiliary variables in calibration estimator using the Shapley decomposition

Statistical Methods & Applications - In multipurpose surveys several interest variables and a very large number of auxiliary variables are collected. Auxiliary variables are usually considered...     

Attachment Representations and Time Perspective in Adolescence

This study examines the relationship between attachment to parents and peers, time perspective and psychological adjustment in adolescence. 2,665 adolescents (M age = 17.03 years, SD = 1.48) completed self-report measures about parent and peer attachment, time perspective, sympathy and self-determination. Subjects were divided into four groups based on their parent and peer attachment scores: high parent and high peer attachment, low peer and low parent, high parent and low peer, low parent and high peer. Higher levels of negative past of the Zimbardo Time Perspective Inventory were revealed, by post hoc analyses, in the group with low attachment to parent and peer. Unlike adolescents with low attachment to parents, adolescents who reported a secure attachment to parents had higher scores on positive past, hedonistic present, and future. Adolescents with high attachment to parent and peer and adolescents with high parent but low peer attachment had the highest scores on competence and authonomy.     

The Recent Crisis Impact on the Economies of Some Countries of Central and Eastern Europe

Wealth of nations is nowadays something complex depending on democratic processes and debt accumulation. In Europe the EURO introduction provoked even more complexity, as there is no common government to steer the fiscal process, a situation without precedence. New solutions are therefore looked for, based on different schools of thought in order to tackle the problem. The analytical part analyzes the economic behaviour of four recently less studied countries. A comparison between them is based on time series of Real GDP Growth, Consumer Price Inflation, Government Balance.     

Prognostic assessment of repeatedly measured time-dependent biomarkers,with application to dilated cardiomyopathy

We propose new time-dependent sensitivity, specificity, ROC curves and net reclassification indices that can take into account biomarkers or scores that are repeatedly measured at different time-points. Inference proceeds through inverse probability weighting and resampling. The newly proposed measures exploit the information contained in biomarkers measured at different visits, rather than using only the measurements at the first visits. The contribution is illustrated via simulations and an original application on patients affected by dilated cardiomiopathy. The aim is to evaluate if repeated binary measurements of right ventricular dysfunction bring additive prognostic information on mortality/urgent heart transplant. It is shown that taking into account the trajectory of the new biomarker improves risk classification, while the first measurement alone might not be sufficiently informative. The methods are implemented in an R package (longROC), freely available on CRAN.     

Modelling work history patterns in the Italian labour market

We focus on work histories of new entrants in 1998 in the Italian labour market. For workers in the private sector, we define a standard and three non-standard history patterns. We profile the workers through a mixed-effect multinomial logit model and show that certain features may be associated with the probability of belonging to one or the other category. Furthermore, we show that there are differential effects on wages associated with non-standard patterns. A closer look at best performing non-standard workers shows that even for them an early contractual stabilization may not always be expected.     

Evaluating the human resource related soft dimensions in green supply chain management implementation

Because of increased carbon emissions, environmental protection initiatives have gained significant attention at global level. One of the major initiatives taken by the industrial sector to minimize the negative environmental effect of the value chain activities is Green Supply Chain Management (GSCM). In industry, soft (human resource-related) dimensions influence the implementation of GSCM process greatly. In the literature, relatively less discussion is provided on assessing the significance of soft dimensions in efficient GSCM acceptance in industry. The present work is an attempt to construct a structural framework for assessing the significance of the soft dimensions in adopting GSCM concepts by taking a case of automotive company in India. A hybrid approach of Best Worst Method (BWM) and Decision-Making Trial and Evaluation Laboratory (DEMATEL) approach is employed in this work. BWM is used to prioritize the GSCM oriented soft dimensions, and DEMATEL is employed to extract interrelationships among them. The result shows that ‘Top management commitment’, ‘Employee involvement’, ‘Organizational culture’ and ‘Teamwork’ are the highly prioritized causal soft dimensions in efficient GSCM adoption. This research work would help industry managers and practitioners to decide where to concentrate for GSCM concepts in context of soft dimensions for sustainable business development.     

Snipping for robust k-means clustering under component-wise contamination

We introduce the concept of snipping, complementing that of trimming, in robust cluster analysis. An observation is snipped when some of its dimensions are discarded, but the remaining are used for clustering and estimation. Snipped k-means is performed through a probabilistic optimization algorithm which is guaranteed to converge to the global optimum. We show global robustness properties of our snipped k-means procedure. Simulations and a real data application to optical recognition of handwritten digits are used to illustrate and compare the approach.