Search

In order to propose a statistical methodology that allows to model asymmetric data using the Lévy distribution, a simulation study is presented under nine different scenarios to evaluate the estimation of the parameters of the distribution in the two approaches of statistics (Classical and Bayesian). The probability distributions Log-Normal, Lévy and Lévy Standard were considered to model the behavior of two real data sets with positive asymmetry, finding that the Lévy distribution fitted well to the proposed data set, therefore the Lévy distribution can be considered as a candidate to adjust asymmetric data with the presence of extreme values. [ABSTRACT FROM AUTHOR]

In this work we develop statistical inference procedures, based on bootstrap and permutation tests, for the European Customer Satisfaction Index (ECSI). The main objective is to fill an existent gap, to the best of our knowledge, in statistical methods for confidence interval estimation of one ECSI, for hypothesis testing comparing two or more ECSIs, and for ranking a set of ECSIs. The procedures proposed is applied to a study of higher education students' satisfaction conducted in Veracruz University (México). The results indicate that the procedures of statistical inference are feasible and straightforward to implement with R. [ABSTRACT FROM AUTHOR]

A class of estimators is determined. It contains important families of ratio estimators. The belonging to it is characterized by means of a parametric vector. It allows identifying that more than a dozen of estimators are particular cases. Their behavior when ranked set sampling is used is established by fixing the gains in accuracy of them. Numerical experiments establish which is the performance of them in four applications and by generating normal, exponential and uniform populations. [ABSTRACT FROM AUTHOR]

Binary logistic regression as an instrument for the prediction of immunological deterioration from nutritional indicators in persons infected with HIV-AIDS. In the evolution of infection with HIV diverse markers of progression have been used, of immunological, virological and genetic type. However, the significance of nutritional indicators for prognosis has been insufficiently studied. The objective of this investigation was to use binary logistic regression as a tool to find a mathematical model which would allow to predict the counting of CD4+ cells in persons infected with HIV-AIDS after a year's follow-up. 118 persons with HIV-AIDS were studied (initial time and after a year of evolution) which were classified in four subgroups according to the use or not use of antirretrovirals and the counting of CD4 at the initial time. Nutritional, biochemical and anthropometric variables were used, as well as the gradients between both observations. Binary logistic regression was applied, having as dependent variable the counting of CD4 as an immunological criterion of antirretroviral therapy. The logistical models found were adjusted according to the subgroup they belonged to, and they were compared to one another according to the goodness of fit and accuracy. The step-by-step method chose only the gradient of body mass index (IMC) as independent variable. The adjustment of the model allowed to propose a mathematical equation by subgroup in order to predict the need of antirretroviral therapy after a year. The groups which had initial values of CD4 < 350, even with an increase in IMC ran the risk of keeping immunological deterioration after a year. All models showed specificity values higher than those of sensitiveness, though models II and III were the best, as proved by the probabilities shown in ROC curves. The conclusion is that the IMC gradient in a year of evolution might be used in a logistical mathematical model to predict the need of antirretroviral therapy in individuals with HIV-AIDS. [ABSTRACT FROM AUTHOR]

To be able to resolve tomographic inversion problems in travel time, it is necessary the construction of distance matrices containing information of survey array geometry, the discretized domain shape and the distance traveled by rays in each area elements that compose it. Domain considered are generally rectangular for ease of geometric resolution. In this work is circular due to the need to carry out investigation in columns of this shape. Distance matrices are used to solve the linear system, which leads to the determination of velocity values in each element of the domain. Taking in mind discretization, there are several possibilities for the construction of arrays of pixels: may those be defined in one quadrant and obtaining the rest by symmetry; build arrays using any size pixels independent of the quadrant, and finally construct distance matrices with square pixels. Each option will provide access to new research possibilities. In this work an original procedure was developed for the construction of matrices with elements of the last type. The procedure was validated with simulated data using the damped least squares solution (DLSS) method, which demonstrated the efficiency of the algorithm. [ABSTRACT FROM AUTHOR]

In this work we have collected estimators of the population proportion through the use of auxiliary information under simple random sampling. We analyzed the main properties of these estimators. Theorethical properties suggest that the estimators with use of auxiliary information can outperform alternative methods, and the results derived from a Monte Carlo simulation study support this view. Has made an application on a population of Risk of Acidication in Lakes in the Northeastern states of the U.S. conducted by the Environmental Monitoring and Assessment Program of the US Environmental Protection Agency. In the application we have compared the different estimators each other with the standard stimators. [ABSTRACT FROM AUTHOR]

In the paper the results of the combined strategy formed by Robust Principal Component Analysis and CUR matrix decomposition are applied to a practical problem for adjusting a regression model by using Partial Least Squares (PLS). The conclusions concerning their advantages of such an approach as well information about the adjusted models are given. [ABSTRACT FROM AUTHOR]

In a classic problem of competing risks, estimates of diverse rates are carried out starting from suppositions about the causes that compete. These suppositions have included essentially the independence among the competing causes. The present work intends to relax these suppositions to more general situations. It is demonstrated that the Net Rate can be estimated from observed data in the presence of Competing Risks with certain type of dependence. Generalizations of the Actuarial estimator for the Net Rate are suggested for the two classes of dependencies studied. By means of simulations, the proposed estimators and the one reported in the literature for independent competing risks are compared, and the results confirm the fact that the new estimators can be used in the circumstances defined to estimate the probabilities of survival. [ABSTRACT FROM AUTHOR]

The technique of Randomized Response has been introduced to reduce the risk of evasive answer or no-response in survey samplings of sensitive issues to estimate the total of individuals with that sensitive characteristic. We also know that highly efficient strategies of estimation require strong auxiliary information. Hence, in this article we obtain and compare two estimators: The Rao-Hartley-Cochran (RHC) estimator with Warner's model and the RHC estimator with Greenberg's model, both in a finite population setting. We will show via simulation that ŧA1,RHC,π the RHC π-estimator in Warner's model is less efficient than ŧA1,RHC,π, the RHC π-estimator in Greenberg's model. [ABSTRACT FROM AUTHOR]

The majority of the developments in a classic problem of competing risks have consisted in, based on the assumption of independence between all event types, make estimations to several rates. In this paper, we propose an adjusted Life-Table estimator in the context of non-independent competing risks. We proof that the proposed estimator is maximum pseudo-likelihood. We compare the adjusted and classic estimators by simulation study. It confirmed that de adjusted estimator provides closer estimations to the real rate than the classic estimator. [ABSTRACT FROM AUTHOR]