Search

Abstract  In this paper, from Frèchet’s metric, diagnostic tools are constructed for the detection of influential observations in Profile Analysis with elliptically distributed random errors . This distributional hypothesis allows the application of the proposed diagnostics to a wide variety of random experiences, not only for data from a multivariate normal distribution but also from other symmetric distributions, commonly used in studies of several sciences. The diagnostics are based on Frèchet’s distance between the distributions of the basic statistics in Profile Analysis, in the postulated model and in the perturbed model obtained by deleting an observation from the sample data. This metric is highly useful since it enables the analysis of the influence on the point estimation and the estimation error. Applications on two data sets are provided. [ABSTRACT FROM AUTHOR]

In this paper a sufficient condition for the identifiability of finite mixtures is given. This condition is less restrictive than Teicher’s condition Teicher H, Ann Math Stat 34:1265–1269 (1963) and therefore it can be applied to a wider range of families of mixtures. In particular, it applies to the classes of all finite mixtures of Log-gamma and of reversed Log-gamma distributions. These families have been already studied by Henna J Jpn Stat Soc 24:193–200 (1994) using another condition, different from Teicher’s, but more difficult to check in many cases. Furthermore, the result given in this paper is very appropiated for the case of mixtures of the union of different distribution families. To illustrate this an application to the class of all finite mixtures generated by the union of Lognormal, Gamma and Weibull distributions is given, where Teicher’s and Henna’s conditions are not applicable [ABSTRACT FROM AUTHOR]

We propose a diagnostic to assess the influence of a unit ui in the sample s, selected from a finite population U={u1,....,uN} using a sampling design D, on θ^(s) as an estimator of a population parameter θ=θ(Y1,...,YN). We adjust the definition of conditional bias (Muñoz-Picardo et al., 1995), in order to obtain an influence diagnostic in survey sampling. Conditional bias as an influence measure of unit ui (ui∈s) on θ^(s), depends on Yi and on the sampling design, D. This is a distinctive feature of sample surveys. Only in particular cases is the proposed influence measure a case-deletion diagnostic. [ABSTRACT FROM PUBLISHER]

The nutritional support is essential in the treatment of the septic patient, but the optimal substrates, have not been well defined, although several studies advise the use of solutions with a high proportion of branched amino acids. The use of lipids in the Parenteral Nutrition (PN) of these patients is presently accepted by all authors, although there is no consensus of the composition of the mix to be used. We have done a prospective, randomized study in a group of critical septic patients who received PN. All cases were used a solution with a high proportion of branched amino acids and glucose. 40% of the non-protein calories were administered as lipids. Based on this, two groups were formed. Group 1 medium chain triglycerides and long chain triglycerides (MCT/LCT) in a ratio of 1:1, and Group II LCT. After ten days, both groups showed an increase in hepatic cholestasis enzymes and an improvement of the parameters of protein anabolism and catabolism. Group 1 showed a significantly higher increase in the retinol birdring protein and the improvement of the nitrogen balance. Besides this, the global analysis of all the parameters of protein anabolism and catabolism showed a more favorable evolution, with statistical significance, in those who received MCT/LCT. Therefore, the MCT/LCT in the PN of the septic patient who is given a solution rich in branched amino acids, achieves a greater protein saving and a faster recevesy of the markers of protein synthesis, than those with LCT. Both groups show an increase in cholestasis enzymes.