Bayesian longitudinal data analysis with mixed models and thick-tailed distributions using MCMC

Linear mixed effects models are frequently used to analyse longitudinal data, due to their flexibility in modelling the covariance structure between and within observations. Further, it is easy to deal with unbalanced data, either with respect to the number of observations per subject or per time period, and with varying time intervals between observations. In most applications of mixed models to biological sciences, a normal distribution is assumed both for the random effects and for the residuals. This, however, makes inferences vulnerable to the presence of outliers. Here, linear mixed models employing thick-tailed distributions for robust inferences in longitudinal data analysis are described. Specific distributions discussed include the Student-t, the slash and the contaminated normal. A Bayesian framework is adopted, and the Gibbs sampler and the Metropolis-Hastings algorithms are used to carry out the posterior analyses. An example with data on orthodontic distance growth in children is discussed to illustrate the methodology. Analyses based on either the Student-t distribution or on the usual Gaussian assumption are contrasted. The thick-tailed distributions provide an appealing robust alternative to the Gaussian process for modelling distributions of the random effects and of residuals in linear mixed models, and the MCMC implementation allows the computations to be performed in a flexible manner.

Basis sets applied to the theoretical study of the vibrational structure of hexaaquaaluminum(III) ion

Gaussian basis sets were developed with the Generator Coordinate Hartree-Fock (GCHF) method for the atoms from H (14s), O (23s16p), and Al (29sl9p) in the ground state. These basis sets were then contracted to 3s (12,1,1), 5s3p (18,2,1,1,1/14,1,1), and 7s5p (20,3,2,1,1,1,1/14,2,1,1,1) for H, O and Al atoms, respectively, by a standard procedure. The quality of contracted basis sets in molecular calculations was evaluated through studies of the total and orbital (epsilon(HOMO) and epsilon(HOMO-1)) energies at the HF level for the hexaaquaaluminum(III) ion, [AI(H(2)O)(6)](3+). For the O atom, the 5s3p was supplemented with d polarization function and it was used in combination with 3s, and 7s5p for H and Al atoms was used to the theoretical interpretation of the Infrared (IR) spectrum of hexaaquaaluminum(III) ion. The calculations of the IR-spectrum were also performed at the HF level and it showed that the basis sets obtained with the aid of GCHF method lead to the selection of useful contracted Gaussian basis sets for the theoretical study of vibrational property of ionic specie of our interest. (C) 2004 Elsevier B.V. All rights reserved.

Integration of polyharmonic functions

The results in this paper are motivated by two analogies. First, m-harmonic functions in R(n) are extensions of the univariate algebraic polynomials of odd degree 2m-1. Second, Gauss' and Pizzetti's mean value formulae are natural multivariate analogues of the rectangular and Taylor's quadrature formulae, respectively. This point of view suggests that some theorems concerning quadrature rules could be generalized to results about integration of polyharmonic functions. This is done for the Tchakaloff-Obrechkoff quadrature formula and for the Gaussian quadrature with two nodes.

Development of basis sets to calculations of the electronic structure of YMnO3

The generator coordinate Hartree-Fock method was used to develop 20s17p, 30s20p14d, and 30s21p16d Gaussian basis sets for the O ((3)p), Mn (S-6), and Y (D-2) atoms, respectively. The Gaussian basis sets were contracted to 20s17p/9s7p, 30s20p14d/11s7p7d, and 30s21p16d/14s7p7d and utilized in calculations of total energy and orbital energies of the (MnO1+)-Mn-5 and (YO1+)-Y-3 fragments to evaluate its quality in molecular studies. Finally, the contracted basis set for O atom was supplemented with one polarization function of d symmetry and used along with the other contracted basis sets (for Mn and Y) to calculate dipole moments, total energy, and total atomic charges in YMnO3 in space group D-6h. The analysis of those properties showed that is reasonable to believe that YMnO3 present behavior of piezoelectric material. (C) 2003 Elsevier B.V. All rights reserved.

Experimental and theoretical study of the compound [Pd(dmba)(NCO)(imz)]

The compound [Pd(dmba)(NCO)(imz)] (dmba = N,N-dimethylbenzilamine; NCO = cyanate; imz = imidazole) was studied through experimental and theoretical methods. The complex was synthesized and characterized by IR and NMR spectroscopy. To an appropriate representation of the molecular environment, Gaussian basis sets for the constituent atoms of the compound were built and, after adequate supplementation with polarization and diffuse functions, they were used to study the molecule. Calculations of electronic and vibrational structure of two possible isomers were carried out, showing that the compound, which contains the NCO GROUP trans to the Pd-C bond, is 4.29 kcal/mol more stable than the analogous one, where the imz ligand is trans to the Pd-C bond. The calculated molecular parameters, bond distances, and bond angles showed that the geometry around the metallic center is square-planar with the cyanate being linear. The theoretical infrared spectrum of C(1) symmetry (electronic state (1)A) is in accordance with the experimental one. It also verified the contribution of Pd (4d(xz) + 4d(yz)) and Pd (4d(xy)) in the HOMO and LUMO orbitals, respectively. (c) 2006 Elsevier B.V. All rights reserved.

An insight into the theoretical investigation of possible piezoelectric effect in samarium titanate (SmTiO3)

Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)

Photopatternable di-ureasil-zirconium oxocluster organic-inorganic hybrids as cost effective integrated optical substrates

Organic-inorganic hybrids containing methacrylic acid (McOH, CH(2)= C(CH(3))COOH)) modified zirconium tetrapropoxide, Zr(OPr(n))(4), classed as di-ureasil-zirconium oxo-cluster hybrids, have been prepared and structurally characterized by X-ray diffraction (XRD), small-angle X-ray scattering (SAXS), Fourier transform infrared (FT-IR) and Raman (FT-Raman) spectroscopies, Si and C nuclear magnetic resonance (NMR), and atomic force microscopy (AFM). XRD and SAXS results have pointed out the presence of Si- and Zr-based nanobuilding blocks (NBBs) dispersed into the organic phase. Inter-NBBs correlation distances have been estimated for the pure di-ureasil and a model compound obtained. by hydrolysis/condensation of Zr(OPr(n))(4):McOH (molar ratio 1: 1): d(Si) approximate to 26 +/- 1 angstrom and d(Zr) approximate to 16 +/- 1 angstrom, respectively. In the case of the di-ureasil-zirconium oxo-cluster hybrids, these distances depend on the Zr relative molar percentage (rel. mol. Zr %) (d(Si) ranges from 18 to 25 angstrom and d(Zr) from 14 to 23 angstrom, as the rel. mol. Zr % increases from 5 to 75), suggesting that the Si- and Zr-based clusters are interconstrained. Complementary data from FT-IR, FT-Raman, (29)Si and (13)C NMR, and AFM support to a structural model where McOH-modified Zr-based NBBs (Zr-OMc) are present over the whole range of composition. At low Zr-OMc contents (rel. mol. Zr % <30) the clusters are well-dispersed within the di-ureasil host, whereas segregation occurs at the 0.1 mu m scale at high Zr-OMc concentration (rel. mol. Zr % = 50). No Zr-O-Si heterocondensation has been discerned. Monomode waveguides, diffractions gratings, and Fabry-Perot cavities have been written through the exposure of the hybrid monoliths to UV light. FT-Raman has shown that the chemical process that takes place under illumination is the polymerization of the methacrylate groups of the Zr-OMc NBBs. The guidance region in patterned channels is a Gaussian section located below the exposed surface with typical dimensions of 320 mu m wide and 88 mu m deep. The effective refractive index is 1.5162 (maximum index contrast on the order of 1 x 10(-4)) and the reflection coeficient of the Fabry-Perot cavity (formed by a grating patterned into a 0.278 cm channel) is 0.042 with a free spectral range value of 35.6 GHz.

Thermoanalytical study of some anti-inflammatory analgesic agents

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

Quantum chemical modeling of perovskite: An investigation of piezoelectricity in ferrite of yttrium

Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)

Ab Initio correlated all electron dirac-fock calculations for Eka-Francium Fluoride (E119F)

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

Photoluminescence in the CaXSr1-xWO4 system at room temperature

Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)

Elicitation of multivariate prior distributions: A nonparametric Bayesian approach

In the context of Bayesian statistical analysis, elicitation is the process of formulating a prior density f(.) about one or more uncertain quantities to represent a person's knowledge and beliefs. Several different methods of eliciting prior distributions for one unknown parameter have been proposed. However, there are relatively few methods for specifying a multivariate prior distribution and most are just applicable to specific classes of problems and/or based on restrictive conditions, such as independence of variables. Besides, many of these procedures require the elicitation of variances and correlations, and sometimes elicitation of hyperparameters which are difficult for experts to specify in practice. Garthwaite et al. (2005) discuss the different methods proposed in the literature and the difficulties of eliciting multivariate prior distributions. We describe a flexible method of eliciting multivariate prior distributions applicable to a wide class of practical problems. Our approach does not assume a parametric form for the unknown prior density f(.), instead we use nonparametric Bayesian inference, modelling f(.) by a Gaussian process prior distribution. The expert is then asked to specify certain summaries of his/her distribution, such as the mean, mode, marginal quantiles and a small number of joint probabilities. The analyst receives that information, treating it as a data set D with which to update his/her prior beliefs to obtain the posterior distribution for f(.). Theoretical properties of joint and marginal priors are derived and numerical illustrations to demonstrate our approach are given. (C) 2010 Elsevier B.V. All rights reserved.

Spatial variability models of CO2 emissions from soils colonized by grass (Deschampsia antarctica) and moss (Sanionia uncinata) in Admiralty Bay, King George Island

Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)

Initial-condition problem for a chiral Gross-Neveu system

A time-dependent projection technique is used to treat the initial-value problem for self-interacting fermionic fields. On the basis of the general dynamics of the fields, we derive formal equations of kinetic-type for the set of one-body dynamical variables. A nonperturbative mean-field expansion can be written for these equations. We treat this expansion in lowest order, which corresponds to the Gaussian mean-field approximation, for a uniform system described by the chiral Gross-Neveu Hamiltonian. Standard stationary features of the model, such as dynamical mass generation due to chiral symmetry breaking and a phenomenon analogous to dimensional transmutation, are reobtained in this context. The mean-field time evolution of nonequilibrium initial states is discussed.

Testing the resolving power of two-dimensional K+K+ interferometry

We show results from an analysis performed to test the resolving power of a two-dimensional χ2 method proposed previously when applied to the case of kaon interferometry, where no significant contribution from long-lived resonances is expected. For that purpose, use is made of the preliminary E859 K+K+ interferometry data from Si+Au collisions at 14.6/4 GeV/c. Although less sensitivity is achieved in the present case, this analysis seems to favor scenarios with no resonance formation at the AGS energy range. The possible compatibility of data with zero decoupling proper time interval, conjectured by the three-dimensional experimental analysis, is also investigated and is ruled out when considering more realistic dynamical models with expanding sources. Furthermore, these results strongly emphasize that the static Gaussian parametrization cannot be trusted under more realistic conditions, leading to a distorted or even wrong interpretation of the source parameters.