1,720,987 research outputs found
Efficient unequal probability resampling from finite populations
Full text linkA resampling technique for probability-proportional-to size sampling designs is proposed. It is essentially based on a special form of variable probability, without replacement sampling applied directly to the sample data, yet according to the pseudo-population approach. From a theoretical point of view, it is asymptotically correct: as both the sample size and the population size increase, under mild regularity conditions the proposed resampling design tends to coincide with the original sampling design under which sample data were collected. From a computational point of view, the proposed methodology is easy to be implemented and efficient, because it neither requires the actual construction of the pseudo-population nor any form of randomization to ensure integer weights and sizes. Empirical evidence based on a simulation study1 indicates that the proposed resampling technique outperforms its two main competitors for confidence interval construction of various population parameters including quantiles. (c) 2021 Published by Elsevier B.V
Comparing Recent Approaches For Bootstrapping Sample Survey Data: A First Step Towards A Unified Approach
No full textBootstrap algorithms are simple and appealing solutions for variance estimation under a complex sampling design, however, they must account for the non-iid nature of data. Literature about boot- strapping finite population samples appears to have developed according to two major approaches. A more practical ad-hoc approach refers to the so-called scaling problem and is based on a data- rescaling so that, in the linear case, the resulting bootstrap estimate for the variance perfectly matches the analytic variance estimate. A more fundamental plug-in approach is based on the mim- icking bootstrap principle and on the bootstrap population created on the basis of (original) sample data. Recent proposals suggest a direct bootstrap matching the linear case variance but avoiding any data scaling under mixed re-sampling designs. In this paper, a new perspective to the bootstrap population plug-in approach is provided that avoids the physical reconstruction of the bootstrap population. Basic sampling designs, both with and without replacement as well as unequal proba- bility designs are considered. Focusing on probability-proportional-to-size sampling, a simulation study is conducted that compares all the approaches considered
Resampling from finite populations: An empirical process approach.
No full textIn sampling finite populations, several resampling schemes have been proposed. The common starting point is that, despite its excellent asymptotic properties, Efron’s original bootstrap only works for i.i.d. data. This condition is not met in sampling finite populations, because of the dependence among units due to the sampling design. Hence, adaptations are needed to account for the non i.i.d. nature of data. Different versions of the standard bootstrap algorithm have been proposed in the literature. A new class of resampling procedures for finite populations is defined. Such a class appears to provide a unified framework that allows for encompassing other resampling algorithms already proposed. Its main theoretical justification is based on asymptotic, large sample arguments: the probability distribution of the original statistic and its approximation based on resampling converge to the same limit. Technically speaking, it is shown that a “finite population version” of the empirical process and its “resampled form” weakly converge to the same limiting Gaussian process. In a sense, this justification is similar to those given for classical bootstrap
Cleaning validation of a fermentation group for the production of biomedical products [Cleaning validation di un gruppo prefermentatore-fermentatore per la produzione di prodotti biomedicali]
No full textThe fermentation-group is largely employed to produce pharmaceuticals by means of biotechnology techniques. The operative functions of these plants are surely kept under control by the validation process, while its cleaning is less investigated. The aim of this work is the cleaning validation of a fermentation-group. Furthermore, Standard Operative Procedures (SOPs) have been validated relating to the production necessities and to the allowed safety limits
A smooth subclass of graphical models for chain graph : towards measuring gender gaps
No full textRecent gender literature shows a growing demand of sound statistical methods for analysing gender gaps, for capturing their complexity and for exploring the pattern of relationships among a collection of observable variables selected in order to disentangle the latent trait of gender equity. In this paper we consider parametric Hierarchical Marginal Models applying to binary and categorical data, as a promising statistical tool for gender studies. We explore the potential of Chain Graphical Models, by focusing on a special smooth sub-class of models known as Graphical Models of type II as recently developed (Nicolussi in Marginal parameterizations for conditional independence models and graphical models for categorical data, 2013) , i.e. an advanced methodology for untangling and highlighting any dependence/independence pattern between gender and a set of covariates of mixed nature, either categorical, ordinal or quantitative. With respect to traditional methodologies for treating categorical variables, such as Logistic Regression and Chi-Squared test for contingency table, the proposed model lead to a full multivariate analysis, allowing for isolating the effect of each dependent variable from all other response variables. At the same time, the resulting graph offers an immediate visual idea of the association pattern in the entire set of study variables. The empirical performance of the method is tested by using data from a recent survey about sexual harassment issues inside university, granted by the Equal Opportunities Committee of the University of Milano-Bicocca (Italy)
A unified principled framework for resampling based on pseudo-populations: asymptotic theory
No full text- …
