University of Extremadura

About: University of Extremadura is a education organization based out in Badajoz, Spain. It is known for research contribution in the topics: Population & Hyperspectral imaging. The organization has 7856 authors who have published 18299 publications receiving 396126 citations. The organization is also known as: Universidad de Extremadura.

A chloroform extract obtained from a decoction of Ficus carica leaves improves the cholesterolaemic status of rats with streptootocin-includede diabetes

Abstract: The antidiabetic effects of Ficus carica leaf extracts have been reported previously From the aqueous decoction of fig leaves, after treatment with HCI, centrifuging, treatment with sodium hydroxide (NaOH) and extraction with chloroform (CHCl3), the administration of the organic phase rats with streptozotocin-induced diabetes led to a decline in the levels of total cholesterol and an decrease in the total cholesterol/HDL cholesterol ratio (with respect to the control group), together with a reduction of the hyperglycaemia

Data-driven density derivative estimation, with applications to nonparametric clustering and bump hunting

Abstract: Important information concerning a multivariate data set, such as clusters and modal regions, is contained in the derivatives of the probability density function. Despite this importance, nonparametric estimation of higher order derivatives of the density functions have received only relatively scant attention. Kernel estimators of density functions are widely used as they exhibit excellent theoretical and practical properties, though their generalization to density derivatives has progressed more slowly due to the mathematical intractabilities encountered in the crucial problem of bandwidth (or smoothing parameter) selection. This paper presents the first fully automatic, data-based bandwidth selectors for multivariate kernel density derivative estimators. This is achieved by synthesizing recent advances in matrix analytic theory which allow mathematically and computationally tractable representations of higher order derivatives of multivariate vector valued functions. The theoretical asymptotic properties as well as the finite sample behaviour of the proposed selectors are studied. In addition, we explore in detail the applications of the new data-driven methods for two other statistical problems: clustering and bump hunting. The introduced techniques are combined with the mean shift algorithm to develop novel automatic, nonparametric clustering procedures which are shown to outperform mixture-model cluster analysis and other recent nonparametric approaches in practice. Furthermore, the advantage of the use of smoothing parameters designed for density derivative estimation for feature significance analysis for bump hunting is illustrated with a real data example.