M. COSTA, International financial flows and regional disparities: a statistical analysis, pp. 583-602
The paper presents a statistical methodology able to evaluate European Monetary Union in effects on international financial flows. Foreign direct investment and portfolio investment, interpreted as indicators of growth expectations, are analyzed in order to detect if Euro introduction represents a source of economic integration among EMU countries. By referring to risk sharing methods, portfolio investment flows are founded to be a risk sharing cannel, while direct investment flows increase consumption variability.
E. FABRIZI, Hierarchical Bayesian models for the estimation of unemployment rates in small domains of the italian labour force survey, pp. 603-618
In this paper we introduce shrinkage estimators for the estimation of the Unemployment rate in small domains of the Italian Labour Force Survey. The proposed estimators are based on Hierarchical Linear Mixed Models and on the borrowing strength on both time series and cross section. Auxiliary information from source external to the Labour Force Survey is not considered. A Hierarchical Bayesian approach is adopted, in which models are solved by means of MCMC sampling algorithms. This allows to measure variability associated to estimators accounting, in a simple way, for all e uncertainty sources. Results highlight how, simple hierarchical models allows for remarkable gain in efficiency with respect to published estimates, and that models with a time series component perform better than those based exclusively on data from the same repetition of the survey.
M. A. W. MAHMOUD, N. A. ABDUL ALIM, On testing exponentiality against NBARFR life distribution, pp.618-630
This paper considers testing exponentiality against new better than average renewal failure rate (NBARFR) alternatives. The percentiles of this test statistic are tabulated for simple size 5(1)50. Pitman's asymptotic efficiencies relative to the tests of the new better than used (NBU), new better than used failure rate (NBUFR) and new better than used renewal failure rate (NBURFR) (Amhmed, 1994; Hendi et al., 2000 and Mahmoud and Abdul Alim, 2002). The powers of this test are also calculated for some used life distributions. The problem when right-censored data is available is handled. Pratical applications of our tests in the medical sciences are present.
P. SILVAPULLE, Testing for seasonal fractional integration in quarterly time series, pp.633-646
Many series such as agricultural commodity prices and economic and financial series, exhibit strong dependence-long memory property.
Since many time series also exhibit seasonal patterns, this paper considers a number of tests - namely Hassler's extension of Geweke and Porter-Hudack's (1983) (GPH) semi-parametric test, Robinson's frequency domain score test and Silvapulle's time domain test - to assess the long memory properties of quarterly time series at zero and sesonal frequencies. Very little is known about the finite sample statistical property of these tests. In a simulation study, we find that time domain and semi-parametric tests generally have the rejection rates under the null hypothesis close to the nominal level, with the latter tests' rejection rates higher than the nominal level at the semi-annual frequency. In terms of power, the time domain score test was shown to be superior with respect to the others. Establishing the reliability of these tests in finite samples is very useful to applied researchers.
R. PRISCO, G. CARAMIA, Frequency anomalies pertaining to Mendel's second experiment, pp.646-660
This article is inspired by the second experiment executed by Mendel on three joint characters (shape, color, seed-coat). Through various experiments he managed to isolate plants which differed by more than one character and noted that they were transmitted indipendently from one another, following a relationship 1:2:1 (dominant: hybrid: recessive). Such relationship togheter with the hypothesis of indipendence, to be verified on the basis of empirical data, represent the basis for estimating the theoretical frequencies. Among the various indeces which may be used to describe the difference between theoretical and observed frequencies, we have chosen G2, which was calculated for each of the nine tables conditioned to the three modalities of every character. The partition by addition of G2 allows the definition of three hypotheses on the characters and a total hypothesis. The latter, defined by the intersection of its components, would seem to be the less supported. We defined as anomalous the situation in which, for a special table the total hypothesis results more acceptable than that of one of the components. We have applied a simulation procedure to 100, 000 tables 3x3x3 constructed with the probability relationships theorized by Mendel. The most interesting result is that anomalies are very frequent, as they regard more than 96% of the examined tables.
R. ARBORETTI GIANCRISTOFARO, M. BOLZAN, Nonparametric methods for multivariate observational studies with confounding factors, pp. 663-680
In this paper we propose a path of analysis for multivariate observational studies in presence of confounding factors. We start from a suitable post-stratification by using the propensity score method in order to provide hypotheses testing for coherent alternatives based on nonparametric techniques.
Finally a comparative simulation study to compare the Rosenbaum test (Rosenbaum, 1995) and the multivariate and multistrata permutation test is also presented, even in the presence of missing abservations. .
R. ARBORETTI GIANCRISTOFARO, Multivariate permutation tests in genetics, pp.681-694
In this paper we provide some statistical results for hypotheses testing in genetics particularly referred to multivariate allelic association studies. An extensive power simulation study is also provided on permutation solutions. .
G. GHILARDI, N. ORSINI, Random intercept linear models and estimators for the evaluation of educational system, pp.695-713
This paper presents two intercept estimators, an ordinary one and an empirical Bayesian one, for a multilevel linear regression model where the intercept is treated as a random variable. These estimators can be used as effectiveness indicators for the evulation of the classes, which may vary in terms of the students' characteristics as well as the problems in the student/teacher relationship. The differences between classes size and the intraclass correlation coefficient value, show that empirical Bayes estimator is more efficient than the ordinary estimator. .
S. MIGLIORATI, L. TERZERA, A new estimate of the population size in centre sampling, pp.715-731
The aim of the paper is to give an estimate of the population size when neither a complete list nor some incomplete lists are available and the populution is partially undetectable. In such a framework the center sampling technique proves an essential tool. A first estimate relies on the likelihood approach assuming the complete detectability of the units. Then, this unrealistic hypothesis is removed introducing a prior distribution on detectability, so that a second estimate is given resorting to integrate likelihood. Finally, the previous estimates are compared through a simulation study.
G. VICARIO, G. BARBATO, G. BRONDINO, An application of the asymptotic theory to a threshold model for the estimate of Martens Hardness, pp.733-744
Hardness measurements have a significant role in mechanical metrology, as they are frequently used to characterise materials properties relevant to industrial processes. A recently introduced method, called Martens Hardness, is based on force and indentation records obtained during a test cycle; the Force/Depth Curve, which describes the indetation pattern, is typically formed by two parts having a zero-point in common. A segmented regression model is proposed in this paper, based on the introduction of a threshold parameter in order to estimate the unknown zero-point. The problem is not trivial, since the relationship between observed force and indentation depth is structural and, moreover, the number of nuisance parameters grows with the number of measured data. The asymptotic likelihood theory leads to an estimate of the unknown parameters of the model. Monte Carlo simulations are resorted to in order to analyse the properties of estimators under different hypotheses about measurement errors, and to etablish the applicability conditions of the method proposed.
M. PRATESI, E. ROCCO, Centre sampling for estimating elusive population size, pp.745-757
The estimation of the size of an elusive population is a problem frequently addressed in many fiels of applications. In the paper a sampling strategy for the estimation of the size is proposed, the properties of the estimator are evaluated in a design-based approach. The estimator is unbiased, admissible and consistent and the design is measurable. The expressions of the variance of the population size estimator and of its sample estimator are also proposed. The strategy is applied to a simulated population.
G. CELANT, G.B. DI MASI, Hermite polynomials expansions for discrete-time nonlinear filtering,, pp.759-769
A finite-dimensional approximation to general discrete-time non linear filtering problems is provided. It consists in a direct approximation to the recursive Bayes formula, based on a Hermite polynomials expansion of the transition density of the signal process. Tha approximation is in the sense of convergence, in a suitable weighted norm, to the conditional density of a signal process given the observations. The choise of the norm is in turn made so as to guarantee also the convergence of the conditional moments as well as to allow the evaluation of an upper bound for the approximation error.
M.MAROZZI, An extension of the ordinary permutation solution of the two-sample location problem, pp.771-778
In this paper, a permutation method for comparing locations of two population has been proposed. Due to the way it has been devised, this test allows a deeper inference on testing the studied system of hypotheses than that allowed by the order methods using exactly the same sample data. Furthermore, simulation results show that, in the considered contexts, it performs practically as well as the optimal parametric solution in terms of power and it maintains its size close to the nominal significance level.
D. DE MARTINI, Pointwise estimate of the power and sample size determination for permutation tets, pp.779-790
A method is presented for the estimation of the power of permutation tests when F is unknown. It is based on the natural plug-in of the empirical distribution in the structure of the statistical test, giving the bootstrap power. The consistency of the permutational bootstrap test is shown. Moreover, to determine the sample size m of permutation tests starting from a pilot sample of size n, the "Mapped Bootstrap" is introduced. The Mapped Bootstrap works for a fixed m and is consistent as the pilot sample size n tends to infinity.