Skovgaard`s adjustment to likelihood ratio tests in exponential family nonlinear models

Likelihood ratio tests can be substantially size distorted in small- and moderate-sized samples. In this paper, we apply Skovgaard`s [Skovgaard, I.M., 2001. Likelihood asymptotics. Scandinavian journal of Statistics 28, 3-321] adjusted likelihood ratio statistic to exponential family nonlinear models. We show that the adjustment term has a simple compact form that can be easily implemented from standard statistical software. The adjusted statistic is approximately distributed as X(2) with high degree of accuracy. It is applicable in wide generality since it allows both the parameter of interest and the nuisance parameter to be vector-valued. Unlike the modified profile likelihood ratio statistic obtained from Cox and Reid [Cox, D.R., Reid, N., 1987. Parameter orthogonality and approximate conditional inference. journal of the Royal Statistical Society B49, 1-39], the adjusted statistic proposed here does not require an orthogonal parameterization. Numerical comparison of likelihood-based tests of varying dispersion favors the test we propose and a Bartlett-corrected version of the modified profile likelihood ratio test recently obtained by Cysneiros and Ferrari [Cysneiros, A.H.M.A., Ferrari, S.L.P., 2006. An improved likelihood ratio test for varying dispersion in exponential family nonlinear models. Statistics and Probability Letters 76 (3), 255-265]. (C) 2008 Elsevier B.V. All rights reserved.

Influence diagnostics in beta regression

We consider the issue of assessing influence of observations in the class of beta regression models, which is useful for modelling random variables that assume values in the standard unit interval and are affected by independent variables. We propose a Cook-like distance and also measures of local influence under different perturbation schemes. Applications using real data are presented. (c) 2008 Elsevier B.V.. All rights reserved.

Molecular characterization of human polyomavirus JC in Brazilian AIDS patients with and without progressive multifocal leukoencephalopathy

Background: JC virus (JCV), the causative agent of progressive multifocal leukoencephalopathy (PML), is classified in 8 different genotypes. Previous reports have suggested a positive association between specific genotypes and PML. Objective: To compare genotypes and adaptive mutations of JCV strains from Brazilian AIDS patients with and without PML. Study design: The VP1 region of JCV was amplified by polymerase chain reaction from cerebrospinal fluid samples from 51 patients with PML and from urine samples of 47 patients with AIDS without central nervous system disease. Genotyping was done by phylogenetic analysis. Amino acid replacement and selection pressures were also investigated. Results: JCV genotype frequency distributions showed that genotypes 2 (32.7%), 1 (26.5%) and 3 (23.5%) were the most prevalent. Genotype 1 had a positive association (p < 0.0001) and genotype 3 showed an inverse association (p < 0.001) with PML. A previously undescribed point mutation at residue 91 (L/I or L/V) and (L/P), non-genotype-associated, was found in 5/49 (10.2%) and 2/47 (4.3%) JCV sequences from PML and non-PML patients, respectively. This mutation was under positive selection only in PML patients. A previously described substitution of T-A in position 128 showed a significant difference between PML and non-PML cases (70% versus 16%, respectively, p < 0.0005). Conclusion: In Brazilian patients with AIDS, JCV genotype 1 showed a strong association with PML (p < 0.0001) and JCV genotype 3 showed an inverse association with PML. The possible association of aminoacids substitution in residues 91 and 128 with PML in patients with AIDS must be further investigated. (C) 2010 Elsevier B.V. All rights reserved.

The C(K, X) spaces for compact metric spaces K and X with a uniformly convex maximal factor

In this paper, we prove that if a Banach space X contains some uniformly convex subspace in certain geometric position, then the C(K, X) spaces of all X-valued continuous functions defined on the compact metric spaces K have exactly the same isomorphism classes that the C(K) spaces. This provides a vector-valued extension of classical results of Bessaga and Pelczynski (1960) [2] and Milutin (1966) [13] on the isomorphic classification of the separable C(K) spaces. As a consequence, we show that if 1 < p < q < infinity then for every infinite countable compact metric spaces K(1), K(2), K(3) and K(4) are equivalent: (a) C(K(1), l(p)) circle plus C(K(2), l(q)) is isomorphic to C(K(3), l(p)) circle plus (K(4), l(q)). (b) C(K(1)) is isomorphic to C(K(3)) and C(K(2)) is isomorphic to C(K(4)). (C) 2011 Elsevier Inc. All rights reserved.

Some restriction tests in a new class of regression models for proportions

The main purpose of this work is to study the behaviour of Skovgaard`s [Skovgaard, I.M., 2001. Likelihood asymptotics. Scandinavian journal of Statistics 28, 3-32] adjusted likelihood ratio statistic in testing simple hypothesis in a new class of regression models proposed here. The proposed class of regression models considers Dirichlet distributed observations, and the parameters that index the Dirichlet distributions are related to covariates and unknown regression coefficients. This class is useful for modelling data consisting of multivariate positive observations summing to one and generalizes the beta regression model described in Vasconcellos and Cribari-Neto [Vasconcellos, K.L.P., Cribari-Neto, F., 2005. Improved maximum likelihood estimation in a new class of beta regression models. Brazilian journal of Probability and Statistics 19,13-31]. We show that, for our model, Skovgaard`s adjusted likelihood ratio statistics have a simple compact form that can be easily implemented in standard statistical software. The adjusted statistic is approximately chi-squared distributed with a high degree of accuracy. Some numerical simulations show that the modified test is more reliable in finite samples than the usual likelihood ratio procedure. An empirical application is also presented and discussed. (C) 2009 Elsevier B.V. All rights reserved.

Adsorption of anionic amphiphilic polyelectrolytes onto amino-terminated solid surfaces

The adsorption behavior of several amphiphilic polyelectrolytes of poly(maleic anhydride-alt-styrene) functionalized with naphthyl and phenyl groups, onto amino-terminated silicon wafer has been studied by means of null- ellipsometry, atomic force microscopy (AFM) and contact angle measurements. The maximum of adsorption, Gamma(plateau), varies with the ionic strength, the polyelectrolyte structure and the chain length. Values of Gamma(plateau) obtained at low and high ionic strengths indicate that the adsorption follows the ""screening-reduced adsorption"" regime. Large aggregates were detected in solution by means of dynamic light scattering and fluorescence measurements. However. AFM indicated the formation of smooth layers and the absence of aggregates. A model based on a two-step adsorption behavior was proposed. In the first one, isolated chains in equilibrium with the aggregates in solution adsorbed onto amino-terminated surface. The adsorption is driven by electrostatic interaction between protonated surface and carboxylate groups. This first layer exposes naphtyl or phenyl groups to the solution. The second layer adsorption is now driven by hydrophobic interaction between surface and chains and exposes carboxylate groups to the medium, which repel the forthcoming chain by electrostatic repulsion. Upon drying some hydrophobic naphtyl or phenyl groups might be oriented to the air, as revealed by contact angle measurements. Such amphiphilic polyelectrolyte layers worked well for the building-up of multilayers with chitosan. (C) 2010 Elsevier Ltd. All rights reserved.

Labile Plasma Iron Generation After Intravenous Iron is Time-dependent and Transitory in Patients Undergoing Chronic Hemodialysis

Iron supplementation in hemodialysis patients is fundamental to erythropoiesis, but may cause harmful effects. We measured oxidative stress using labile plasma iron (LPI) after parenteral iron replacement in chronic hemodialysis patients. Intravenous iron saccharate (100 mg) was administered in patients undergoing chronic hemodialysis (N = 20). LPI was measured by an oxidant-sensitive fluorescent probe at the beginning of dialysis session (T0), at 10 min (T1), 20 min (T2), and 30 min (T3) after the infusion of iron and at the subsequent session; P < 0.05 was significant. The LPI values were significantly raised according to the time of administration and were transitory: -0.02 +/- 0.20 mu mol/L at the beginning of the first session, 0.01 +/- 0.26 mu mol/L at T0, 0.03 +/- 0.23 mu mol/L at T1, 0.09 +/- 0.28 mmol/L at T2, 0.18 +/- 0.52 mmol/L at T3, and -0.02 +/- 0.16 mmol/L (P = 0.001 to 0.041) at the beginning of the second session. The LPI level in patients without iron supplementation was -0.06 +/- 0.16 mmol/L. Correlations of LPI according to time were T1, T2, and T3 vs. serum iron (P = 0.01, P = 0.007, and P = 0.0025, respectively), and T2 and T3 vs. transferrin saturation (P = 0.001 and P = 0.0003, respectively). LPI generation after intravenous saccharate administration is time-dependent and transitorily detected during hemodialysis. The LPI increment had a positive correlation to iron and transferrin saturation.