Abstract: In this paper, we study some dynamic generalized information measures between a true distribution and an observed (weighted) distribution, useful in life length studies. Further, some bounds and inequalities related to these measures are also studied.

Abstract: As part of the analysis of the external effectiveness of university education, a special area of attention is represented by graduates’ mobility for occupational reasons Understanding whether or not the various types of degree courses affect mobility to a significant extent and estimating the net effect induced by individual and context-related characteristics as well as the tendency of this phenomenon over time will help provide information support to universities for use in their decision-making processes This article proposes a multi-level longitudinal analysis to investigate the above aspects The analysis draws upon the ALMALAUREA database relative to graduates from the years 2000, 2001 and 2002 interviewed at 1, 3 and 5 years from graduation

Abstract: Corrado Gini (1884-1964) may be considered the greatest Italian statistician. We believe that his important contributions to statistics, however mainly limited to the univariate context, may be profitably employed in modern multivariate statistical methods, aimed at overcoming the curse of dimensionality by decomposing multivariate problems into a series of suitably posed univariate ones. In this paper we critically summarize Gini’s proposals and consider their impact on multivariate statistical methods, both reviewing already well established applications and suggesting new perspectives. Particular attention will be devoted to classification and regression trees, multiple linear regression, linear dimension reduction methods and transvariation based discrimination.

Abstract: In this paper we discuss the use of dummy variables as sensible covariates in a class of statistical models which aim at explaining the subjects’ preferences with respect to several items. After a brief introduction to CUB models, the work considers statistical interpretations of dummy covariates. Then, a simulation study is performed to evaluate the power discrimination of an asymptotic test among sub-populations. Some empirical evidences and concluding remarks end the paper.

Abstract: An important problem in microarray experiments is the detection of genes that are differentially expressed in agiven mumber of classes. We consider a straightforward and easily implemented method for estimating the posterior probability that an individual gene is null. The problem can be expressed in a two-component mixture framework, using an empirical Bayes approach. Current methods of implementing this approach either have some limitations due to the minimal assumptions made or with more specific assumptions are computationally intensive. By converting to a z-score the value of the test statistic used to test the significance of each gene, we can use a simple two-component normal mixture to model adequately the distribution of this score. In the context of the application of this approach to a well known breast cancer data set, we consider some of the issues associated with the problem of the detection of differential expression, including the case where there is need for the use of an empirical null distribution in place of the standard normal (the theoretical null) and the case where none of the genes might be differentially expressed. We also describe briefly some initial results on a cluster analysis approach to this problem, which attempts to model the joint distribution of the individual gene expressions. This latter approach thus has to make distributional assumptions which are note necessary with the former approach based on the z-scores. However, in the case where the distributional assumptions are valid, it has the potential to provide a more powerful analysis.

Abstract: Timeliness is a driving feature of national economic statistics, especially in a short-term frame. In a survey sampling context, the current practice normally consists in a data release process based on a first preliminary estimate available for users within a short-time, followed by a final estimate, available when the data capturing process is considered completed. The number of preliminary estimates can be higher than one: for each of them the magnitude of revisions can be evaluated, on the basis of the difference respect to the final estimate. In this context, according to a model based approach, we propose and compare some preliminary estimation techniques aimed at reducing the average revision. After the definition of the optimal preliminary estimation strategy when the potential non-response bias is ignored, the case when potential differences between preliminary and late respondents can not be neglected is considered as well, with the proposal of a particular poststratification procedure. Further, an empirical comparison among various provisional estimation strategies has been carried out on the basis of the quarterly wholesale trade survey carried out by ISTAT (Italian National Statistical Institute) for the period 2003-2006, aimed at estimating quarterly changes of the average turnover. Results show that a proper model specification leads to preliminary estimation techniques characterised by an average revision lower than that got using the actual respondents’ sample mean.

Abstract: In this paper, we explore the internationalization pattern of firms and its relationship with firms’ heterogeneity. Besides the more traditional exports and Foreign Direct Investments (FDI), we consider various forms of non-equity internationalization. The use of a Multivariate Probit Model allows us to assess the associations among the choices driving the firms’ internationalization strategy as a whole and, at the same time, to avoid a priori assumptions on the internationalization patterns. From the empirical evidence, two main results emerge. At first, we observe that Italian firms jointly adopt various internalization forms, others than exports and FDI, conditionally to characteristics of the firms. The hypothesis reported in literature of a complementary or subsidiary relationship between exports and FDI is then confirmed also for non-equity internationalization forms. Secondly, we find that the heterogeneity of firms, measured by a large range of variables, has an important role in defining the choice of firms on the patterns of internalization. Thus in this context, we endorse the emerging opinion asserting that various dimensions other than productivity are relevant

Abstract: The growing relevance of tourism industry for modern advanced economies has increased the interest among researchers and policy makers in the statistical analysis of destination competitiveness. In this paper we outline a new model of destination competitiveness based on sound theoretical grounds and we develop a statistical test of the model on sample data based on Italian tourist destination decisions and choices. Our model focuses on the tourism decision process which starts from the demand schedule for holidays and ends with the choice of a specific holiday destination. The demand schedule is a function of individual preferences and of destination positioning, while the final decision is a function of the initial demand schedule and the information concerning services for accommodation and recreation in the selected destinations. Moreover, we extend previous studies that focused on image or attributes (such as climate and scenery) by paying more attention to the services for accommodation and recreation in the holiday destinations. We test the proposed model using empirical data collected from a sample of 1.200 Italian tourists interviewed in 2007 (October - December). Data analysis shows that the selection probability for the destination included in the consideration set is not proportional to the share of inclusion because the share of inclusion is determined by the brand image, while the selection of the effective holiday destination is influenced by the real supply conditions. The analysis of Italian tourists preferences underline the existence of a latent demand for foreign holidays which points out a risk of market share reduction for Italian tourism system in the global market. We also find a snow ball effect which helps the most popular destinations, mainly in the northern Italian regions.

Abstract: This paper considers the problem of estimating the reciprocal of the mean of the Inverse Gaussian distribution when a prior estimate or guessed value λ0 of the shape parameter λ is available. We have proposed a class of estimators with its mean squared error formula. Realistic conditions are obtained in which the estimator is better than usual estimator, uniformly minimum variance unbiased estimator (UMVUE) and the minimum mean squared error estimator (MMSE). Numerical illustrations are given in support of the present study.

Abstract: The aim of this paper is to present a consistent estimator of parameters in simultaneous equation model, based on characteristic roots and vectors of a matrix derived from the so called over-identifying restrictions. The Least Orthogonal Distance Estimator presented here is a more recent development of its original limited information version. The occasion, for reviewing it, has been given by its extension to a full information context which is here completely formalized and by the very encouraging results of recent simulation experiments.

Abstract: This work consists of an inferential procedure that allows for a solution to the problem of hypothesis testing, in which the objective is that of comparing the heterogeneity of two populations on the basis of sampling data, ie to test the hypothesis that the heterogeneity of one population is greater or not equal than that of another The simulation study ighlights the good behaviour of the tests, ie the proposed tests are well approximated and powerful

Abstract: This paper proposes duals to Mohanty and Sahoo’s (1995) estimators and analyzes their properties. Unbiased estimators have also been obtained for interpenetrating subsample design and by using Jackknife technique given by Quenouille (1956). An empirical study is carried out to demonstrate the performances of the suggested estimators over other estimators.

Abstract: This paper deals with theoretical concepts and practical examples, aimed at showing that non-Bayesian inference is liable to result in mistakes or unacceptable conclusions, and proves that they are not justified. Section 2 comments on examples when an objective prior distribution exists, and shows how widely one can be mistaken in using a prior quite distant from the real one. Section 3 comments on two results by Godambe, stressing that – in sampling from finite populations – no flat likelihood exists, while an unbiased linear “estimator” with zero variance does not exist, unless we reach a complete knowledge of the population. Section 4 stresses the fundamental difference between a “probability interval” for a parameter, and a “confidence interval” aimed at making inference on the parameter, thus summarizing all certain facts and constraints able to shrink such an inferential interval. Section 5 explains why we are justified in attaching an inductive meaning to a realized confidence interval. Finally, Section 6 counters some well known counter-examples spread in the Bayesian literature, showing that they are unacceptable from a sound inductive basis.

Abstract: In this present study we are going to avail ourselves of figures regarding new employment and employment termination, registered at the Bologna Provincial Labour Exchange for the three-year period 2004-2006, in order to calculate the duration of job placements, according to the type of contract, and the likelihood of temporary workers being made unemployed: in order to do so, we shall utilise the survival curves method based on the Kaplan and Maier filter (Cox and Oates, 1984). Given the impossibility of estimating true transition matrices, in that the database fails to “cover” all outgoing events, the survival curves method at least enables us to estimate the “duration” of permanence in a given state. The utilisation of the Bologna Provincial Employment Centre’s records enables us to cover a sufficiently long period of time, which in turn enables us to obtain sufficiently stable estimates unaffected by contingencies. Clearly, the results may not be generalised for the whole of Italy, although they do nevertheless provide a meaningful insight into the situation of temporary workers (also given the healthy state of the Bologna province’s labour market).

Abstract: Disease mapping studies have been widely performed at univariate level, that is considering only one disease in the estimated models. Nonetheless, simultaneous modeling of different diseases can be a valuable tool both from the epidemiological and from the statistical point of view. In this paper we propose a model for bivariate disease mapping that generalises the univariate CAR distribution. The proposed model is proven to be an effective alternative to existing bivariate models, mainly because it overcome some restrictive hypotheses underlying models previously proposed in this context. Model performances are checked via a simulation study and via application to some real case studies.

Abstract: The density of the likelihood ratio criterion U p , m , n is expressed in terms of a marginal density of a generalized Dirichlet model having a specific set of parameters. The exact distribution of the likelihood ratio criterion so obtained has a very simple and general format for every p . It provides an easy and direct method of computation of the exact p -value of U p , m , n . Various types of properties and relations involving hypergeometric series are also established.

Abstract: In this article, some recurrence relations of inverse and ratio moments for generalized order statistics from doubly truncated and non-truncated generalized exponential distribution are derived. From our results, we deduce the recurrence relations for single and product moments of generalized order statistics from general class distribution obtained by Haseeb and Hassan (2004), also we deduce the recurrence relations for single and product moments of order statistics from generalized exponential distribution obtained by Saran and Pushkarna (2000).

Abstract: In this paper, we shall make use of the properties of the k -th upper record values to develop the inferential procedures such as point estimation. We shall obtain the best linear unbiased estimates of parameters of a two-parameter rectangular distribution based on k -th record values. The efficiency of the best linear unbiased estimates of the parameters based on two k -th record values has been compared with that of the corresponding estimates based on a complete set of k -th record values. At the end we give the characterization of the two-parameter rectangular distribution using k -th upper record values.

Abstract: We show that kernel density estimates of bandwidth h=h(n)→0 satisfy the Cornish-Fisher assumption with parameter m=nh. This allows Cornish-Fisher expansions about the normal for standardized and Studentized kernel density estimates. The expansions given are formal and the conditions for existence/validity are not explored. The expansions lead to first order confidence intervals (CIs) of level 1−ω +O(n−β), where β =p/(2p+ 2) for one-sided CIs and β = p/(p+1) for two-sided CIs, where p is the order of the kernel used. The second order one- and two-sided CIs are given with β =2p/(2p+3) and β =2p/(p+2). We show how to choose the bandwidth for asymptotic optimality.

Abstract: The survey carried out in late 2004 in the Province of Bologna made possible an analysis of the main characteristics of holidays taken in the two-year period 2003-2004. In particular, we evaluated the effects of several sociodemographic traits on the propensity of holidaymaking, taking longer holidays, and starting holidays in peak season. The analyses were developed using multivariate models including Heckman models based on estimation of simultaneous equations in order to take into account the selection bias embedded in the phenomenon under study. Results highlight that tourist behaviours are strongly affected by personal and unobserved attitudes. However, area of residence and occupational status (to be a worker or non-worker, and working category) emerge as key factors that influence holidaymaking propensity and holiday length.

Abstract: Ensemble Prediction Systems play today a fundamental role in weather forecasting. They can represent and measure uncertainty, thereby allowing distributional forecasting as well as deterministic-style forecasts. In this context, we show how the Joint Calibration Model (Agati et al., 2007) – based on a modelization of the Probability Integral Transform distribution – can provide a solution to the problem of information combining in probabilistic forecasting of continuous variables. A case study is presented, where the potentialities of the method are explored and the accuracy of deterministic-style forecasts from JCM is compared with that from Bayesian Model Averaging (Raftery et al., 2005).

Abstract: In large-scale biomolecular sysrems there are frequency distribuions with properties like Stable Laws. It is of interest to construct such frequency distributions. In the present article we consider Cauchy stable law. The large-sample distribution of the Maximum Likelihood Estimator (M.L.E.) of the scale parameter for some discrete distributions generated by Cauchy stable law are investigated. The existence, strong consistency, asymptotic normality and asymptotic efficiency of that is established.