Improving the causal treatment effect estimation with propensity scores by the bootstrap

AStA Advances in Statistical Analysis - When observational studies are used to establish the causal effects of treatments, the estimated effect is affected by treatment selection bias. The inverse...     

Object-Oriented Bayesian Networks for Modeling the Respondent Measurement Error

In this article, Object-Oriented Bayesian Networks (OOBN) are proposed as a tool to model measurement errors in a categorical variable due to respondent. A mixed measurement error model is presented and an OOBN implementing such a model is introduced. The insertion of evidence represented by the observed value and its propagation throughout the network yields for each unit the probability distribution of the true value given the observed. Two methods are used to predict the individual true value and their performance is evaluated via simulation.     

Measuring the impact of the Web: Rasch modelling for survey evaluation

In 2012, the World Wide Web Foundation launched for the first time the Web Index (WI), which combines the existing secondary data with new primary data to rank countries according to their progress and use of the Web. Primary data are gathered via a multi-country specifically designed questionnaire. The aim of our analysis is (1) to evaluate the measurement properties of the expert assessment survey and to provide survey designers with some insights into possible problematic questions and/or unexpectedly behaving countries and (2) to assess the experts’ perception of the state and the value of the Web. To do so the Rating Scale Rasch model is employed. Results show that about 10% of survey questions are detected as misfitting and need to be reworded. Possible reasons are: counter-orientation with respect to the WI polarity, difficulty in understanding the question's words or binary instead of the multiple response scale. Country analysis shows that no country can be considered as an outlier due to notable unexpected pattern of answers. Since the survey is to be expanded in future editions of the WI, the results of our analysis are very important in pin-pointing the questions most in need of refinement for the next edition of the Index.     

Schengen's Soft Underbelly? Irregular Migration and Human Smuggling across Land and Sea Borders to Italy

La migration illégale à destination de l'Italie ou passant par ce pays est généralement associée dans l'esprit des gens à l'image de centaines, voire de milliers de migrants traversant la Méditerranée entassés dans de vieux rafiots se maintenant tout juste à flot. On considère souvent que les traversées de ce type sont la preuve de l'existence de cartels criminels qui seraient structurés hiérarchiquement, centralisés, extrêmement bien organisés et actifs au niveau mondial. Le seul problème avec de telles idées reçues est qu'elles ne concordent absolument pas avec les faits dont on dispose et n'ont pas grand chose à voir avec la réalité concrète des systèmes de migration illégale. Cette étude passe en revue un certain nombre de dossiers de tribunaux italiens concernant diverses organisations qui opèrent aussi bien au travers des frontières terrestres que maritimes. Elle montre comment de telles organisations opèrent en tenant compte de diverses spécificités locales et contraintes structurelles.     

Falsification or choice among alternatives: The unsolved dilemma of hypothesis testing

Summary The scientific attitude towards statistical method has always pursued two basic objectives: identifying false assumptions and selecting, amongst the likely assertions, those which are most consistent with a given system. The methodological demarcation between rejection of a statistical statement, because it is ?false?, or exclusion, because it is ?least probable?, lies in the fundamental premises of inferential procedures. In the first class we find the methods proposed by Fisher, Neyman and Pearson; in the second one, the Bayesian techniques. Even if different inferential theories may coexist, any particular solution has a limit of validity strictly bouded, to the conventional procedural rules on which it is based. Invited paper at the Conference on ?Statistical Tests: Methodology and Econometric Applications?, held in Bologna, Italy, 27–28 May 1993.     

Quantile estimation in ultra-high frequency financial data: a comparison between parametric and semiparametric approach

In the context of ACD models for ultra-high frequency data different specifications are available to estimate the conditional mean of intertrade durations, while quantiles estimation has been completely neglected by literature, even if to trading extent it can be more informative. The main problem arising with quantiles estimation is the correct specification of durations probability law: the usual assumption of Exponentially distributed residuals, is very robust for the estimation of parameters of the conditional mean, but dramatically fails the distributional fit. In this paper a semiparametric approach is formalized, and compared with the parametric one, deriving from Exponential assumption. Empirical evidence for a stock of Italian financial market strongly supports the former approach.Paola Zuccolotto: The author wishes to thank Prof. A. Mazzali, Dott. G. De Luca, Dott. M. Sandri for valuable comments.