20 resultados para Mixed-effect models
em Repositório Institucional UNESP - Universidade Estadual Paulista "Julio de Mesquita Filho"
Resumo:
Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)
Resumo:
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.
Resumo:
Linear mixed effects models have been widely used in analysis of data where responses are clustered around some random effects, so it is not reasonable to assume independence between observations in the same cluster. In most biological applications, it is assumed that the distributions of the random effects and of the residuals are Gaussian. This makes inferences vulnerable to the presence of outliers. Here, linear mixed effects models with normal/independent residual distributions for robust inferences are described. Specific distributions examined include univariate and multivariate versions of the Student-t, the slash and the contaminated normal. A Bayesian framework is adopted and Markov chain Monte Carlo is used to carry out the posterior analysis. The procedures are illustrated using birth weight data on rats in a texicological experiment. Results from the Gaussian and robust models are contrasted, and it is shown how the implementation can be used for outlier detection. The thick-tailed distributions provide an appealing robust alternative to the Gaussian process in linear mixed models, and they are easily implemented using data augmentation and MCMC techniques.
Resumo:
O modelo misto consiste numa importante classe de modelos que tem sido tradicionalmente analisada por meio de procedimentos da análise de variância. Nos modelos mistos, três aspectos são fundamentais: estimação e testes de hipóteses dos efeitos fixos, predição dos efeitos aleatórios e estimação dos componentes de variância. Na análise de modelos lineares mistos desbalanceados, a estimação dos componentes de variância é de fundamental importância e depende da estrutura de covariâncias e dos métodos de estimação utilizados. Nesse contexto, este artigo pretende apresentar os principais métodos de estimação e de análise utilizados no estudo de modelos lineares mistos com estruturas gerais de covariâncias nos efeitos aleatórios, disponíveis no procedimento MIXED, do SAS (Statistical Analysis System).
Resumo:
Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)
Resumo:
In this paper we present a mixed integer model that integrates lot sizing and lot scheduling decisions for the production planning of a soft drink company. The main contribution of the paper is to present a model that differ from others in the literature for the constraints related to the scheduling decisions. The proposed strategy is compared to other strategies presented in the literature.
Resumo:
In this study, we estimated the heritability (h(2)) of earnings in the Quarter Horse in order to evaluate the inclusion of this trait in breeding programs. Records from 14,754 races of 2443 horses from 1978-2009 were provided by Sorocaba Hippodrome, Sao Paulo, Brazil. All ancestors of the registered horses were included in the pedigree file until the 4th generation. Log-transformed performance measures (LPM) were analyzed for animals aged 2, 3, and 4 years and during their entire career. The h(2) estimates were obtained using a multi-trait model and Gibbs sampling that included the effects of sex, year of race, and animal in all analyses. Five analyses were performed: 1 in which LPM was divided by the number of prizes, 1 in which LPM was divided by the number of race starts, and 3 analyses that included the number of prizes, number of race starts, and both (LPM_cNPS) as covariates. Analysis was performed with and without inclusion of the maternal effect. Models were compared based on the deviance information criterion and LPM_cNPS including maternal effects was found to be the best model. The h(2) estimates and standard deviation obtained using model LPM_cNPS were 0.19 +/- 0.08, 0.21 +/- 0.08, 0.22 +/- 0.09, and 0.21 +/- 0.07 for earnings at 2, 3, and 4 years of age and total career, respectively. Our analyses indicate that earnings are subject to selection and can be included in breeding programs to improve the racing performance of Quarter Horses.
Resumo:
Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)
Resumo:
The objective of this study was to evaluate the use of probit and logit link functions for the genetic evaluation of early pregnancy using simulated data. The following simulation/analysis structures were constructed: logit/logit, logit/probit, probit/logit, and probit/probit. The percentages of precocious females were 5, 10, 15, 20, 25 and 30% and were adjusted based on a change in the mean of the latent variable. The parametric heritability (h²) was 0.40. Simulation and genetic evaluation were implemented in the R software. Heritability estimates (ĥ²) were compared with h² using the mean squared error. Pearson correlations between predicted and true breeding values and the percentage of coincidence between true and predicted ranking, considering the 10% of bulls with the highest breeding values (TOP10) were calculated. The mean ĥ² values were under- and overestimated for all percentages of precocious females when logit/probit and probit/logit models used. In addition, the mean squared errors of these models were high when compared with those obtained with the probit/probit and logit/logit models. Considering ĥ², probit/probit and logit/logit were also superior to logit/probit and probit/logit, providing values close to the parametric heritability. Logit/probit and probit/logit presented low Pearson correlations, whereas the correlations obtained with probit/probit and logit/logit ranged from moderate to high. With respect to the TOP10 bulls, logit/probit and probit/logit presented much lower percentages than probit/probit and logit/logit. The genetic parameter estimates and predictions of breeding values of the animals obtained with the logit/logit and probit/probit models were similar. In contrast, the results obtained with probit/logit and logit/probit were not satisfactory. There is need to compare the estimation and prediction ability of logit and probit link functions.
Resumo:
Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)
Resumo:
Plant extracts are some of the most attractive sources of new drugs and have shown promising results for the treatment of gastric ulcers. Several folk medicinal plants and herbs have been used to treat gastrointestinal disorders, including gastric ulcers. Mammea americana L. (Guttiferae) fruit is very common in the diet of the population of northern South America. Our research interest in this plant arose because of its potential medicinal value as a tonic and against stomachache, as used in folk medicine. In this paper we evaluated three different extracts (ethanolic/EtOH, methanolic/MeOH and dichloromethane/DCM) obtained from M. americana L., for their ability to protect the gastric mucosa against injuries caused by necrotizing agents (0.3 M HCI/60% EtOH), hypothermic restraint stress, nonsteroidal anti-inflammatory drugs (NSAID, indomethacin) and pylorus ligation. In the HCI/EtOH-induced gastriculcer model, EtOH and DCM extracts demonstrated significant inhibition of the ulcerative lesion index by 54% (12.0 +/- 2.6 mm) and 86% (3.7 +/- 1.8 mm), respectively, in relation to the control value (26.0 +/- 1.4 mm) (p < 0.000 1). In the NSAID/cholinomimetic-induced lesion model, both EtOH and DCM extracts showed antiulcerogenic effects with significant reduction in the damage to these gastric lesions of 36% (8.3 +/- 2.0 mm) and 42% (7.5 +/- 1.4 mm), respectively, as compared to the control group (13.0 +/- 0.9 mm) (p < 0.0001). In the gastric ulcer induced by hypothermic-restraint stress, both extracts also showed significant activity, and inhibited the gastric lesion index by 58% and 75%, respectively. The EtOH and DCM extracts also changed gastric juice parameters as well as those of cimetidine, decreased gastric acid secretion significantly (p < 0.0001), increased pH values and promoted reduced acid output (p < 0.0001). In all gastric-ulcer-induced models, MeOH extract did not show any significant antiulcerogenic activity, nor did it change gastric-juice parameters (p > 0.05). The results suggest that EtOH and DCM extracts obtained from M. americana possess excellent antisecretory and/or gastrotective effect in all gastric ulcer models. These results suggest that the antiulcerogenic compound(s) present in M. americana may be clustered in the apolar fraction, which will be investigated by our group for the probable mechanisms of action. (c) 2004 Elsevier GmbH. All rights reserved.
Resumo:
The inclusion of the corona effect in a frequency dependent transmission line model is proposed in this paper. The transmission line is represented through a cascade of π circuits and the frequency dependence of the longitudinal parameters is synthesized with series and parallel resistors and inductors. The corona effect will be represented using the Gary and Skilling-Umoto models. The currents and voltages along the line are calculated by using state-space technique. To demonstrate the accuracy and validity of the proposed frequency dependent line model, time domain simulations of a 10 km length single-phase line response under energization procedure will be presented. ©2008 IEEE.
Resumo:
The aim of this study was to compare biofi lm formation by Candida glabrata and Candida albicans on acrylic, either individually or when combined (single and dual species) and then examine the antimicrobial effects of silver nanoparticles and nystatin on these biofi lms. Candidal adhesion and biofi lm assays were performed on acrylic surface in the presence of artifi cial saliva (AS) for 2 h and 48 h, respectively. Candida glabrata and C. albicans adherence was determined by the number of colony forming units (CFUs) recovered from the biofi lms on CHROMagar ® Candida . In addition, crystal violet (CV) staining was used as an indicator of biofi lm biomass and to quantify biofi lm formation ability. Pre-formed biofi lms were treated either with silver nanoparticles or nystatin and the effect of these agents on the biofi lms was evaluated after 24 h. Results showed that both species adhered to and formed biofi lms on acrylic surfaces. A signifi cantly ( P < 0.05) higher number of CFUs was evident in C. glabrata biofi lms compared with those formed by C. albicans . Comparing single and dual species biofi lms, equivalent CFU numbers were evident for the individual species. Both silver nanoparticles and nystatin reduced biofi lm biomass and the CFUs of single and dual species biofi lms ( P < 0.05). Silver nanoparticles had a signifi cantly ( P < 0.05) greater effect on reducing C. glabrata biofi lm biomass compared with C. albicans . Similarly, nystatin was more effective in reducing the number of CFUs of dual species biofi lms compared with those of single species ( P < 0.05). In summary, C. glabrata and C. albicans can co-exist in biofi lms without apparent antagonism, and both silver nanoparticles and nystatin exhibit inhibitory effects on biofi lms of these species. © 2013 ISHAM.
Resumo:
We solve the three-body bound-state problem in three dimensions for mass imbalanced systems of two identical bosons and a third particle in the universal limit where the interactions are assumed to be of zero range. The system displays the Efimov effect and we use the momentum-space wave equation to derive formulas for the scaling factor of the Efimov spectrum for any mass ratio assuming either that two or three of the two-body subsystems have a bound state at zero energy. We consider the single-particle momentum distribution analytically and numerically and analyze the tail of the momentum distribution to obtain the three-body contact parameter. Our findings demonstrate that the functional form of the three-body contact term depends on the mass ratio, and we obtain an analytic expression for this behavior. To exemplify our results, we consider mixtures of lithium with either two caesium or rubidium atoms which are systems of current experimental interest. © 2013 American Physical Society.
Resumo:
Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)