A5044477391- Genetic and phenotypic traits in livestock
- Statistical Methods and Bayesian Inference
- Bayesian Methods and Mixture Models
- Optimal Experimental Design Methods
- Reproductive Physiology in Livestock
- Milk Quality and Mastitis in Dairy Cows
- Genetics and Plant Breeding
- Statistical Methods and Inference
- Advanced Statistical Methods and Models
- Animal Behavior and Welfare Studies
- Mass Spectrometry Techniques and Applications
- Genetic Mapping and Diversity in Plants and Animals
- Isotope Analysis in Ecology
- Sensory Analysis and Statistical Methods
- Gene expression and cancer classification
- Plant Taxonomy and Phylogenetics
- Wheat and Barley Genetics and Pathology
- Statistical Methods and Applications
- Plant Molecular Biology Research
Universidade Estadual de Campinas (UNICAMP)
2011-2024
University of Wisconsin–Madison
2007-2008
Abstract Dark spots in the fleece area are often associated with dark fibres wool, which limits its competitiveness other textile fibres. Field data from a sheep experiment Uruguay revealed an excess number of zeros for spots. We compared performance four Poisson and zero-inflated (ZIP) models under simulation scenarios. All performed reasonably well same scenario were simulated. The deviance information criterion favoured model residual, while ZIP residual gave estimates closer to their...
Summary Mastitis in cows can be defined as a binary trait, reflecting presence or absence of clinical mastitis (CM), count variable, number cases (NCM), within time interval. Many different models have been proposed for genetic analyses mastitis, and the objective this study was to evaluate predictive ability sire predictions set evaluation CM NCM. Linear‐ threshold liability CM, linear, censored ordinal threshold, zero‐inflated Poisson (ZIP) NCM were compared cross‐validation study. To...
In this research article, we propose a class of models for positive and zero responses by means zero‐augmented mixed regression model. Under class, are particularly interested in studying whose distribution accommodates skewness. At the same time, can be zero, therefore, justify use mixture We model mean response logarithmic scale probability logit scale, both as function fixed random effects. Moreover, effects link two components through their joint incorporate within‐subject correlation...
In this study, we deal with the problem of overdispersion beyond extra zeros for a collection counts that can be correlated. Poisson, negative binomial, zero-inflated Poisson and binomial distributions have been considered. First, propose multivariate count model in which all follow same distribution are Then extend sense correlated may different distributions. To accommodate correlation among counts, considered random effects each individual mean structure, thus inducing dependency common...
Response variables that are scored as counts, for example, number of mastitis cases in dairy cattle, often arise quantitative genetic analysis. When the zeros exceeds amount expected such under Poisson density, zero-inflated (ZIP) model is more appropriate. In using ZIP animal breeding studies, it necessary to accommodate and environmental covariances. For that, this study proposes mixture parameters hierarchically, each a function two random effects, representing sources variability,...
Abstract In this paper, we propose a model based on class of symmetric distributions, which avoids the transformation data, stabilizes variance observations, and provides robust estimation parameters high flexibility for modeling different types data. Probabilistic statistical aspects new are developed throughout article, include mathematical properties, inference. The obtained results illustrated by means real genomic Keywords: EM algorithmJohnson system distributionsmaximum-likelihood...
Abstract This article is motivated by the challenge of analysing an agricultural field experiment with observations that are positive on a continuous scale or zero. Such data can be analysed using two-part models, where distribution mixture and Bernoulli distribution. However, traditional models do not include any dependencies between two parts model. Since probability zero anticipated to high when expected value part low, other way around, this introduces dependency-extended models. In...
We consider unsupervised classification by means of a latent multinomial variable which categorizes scalar response into one the L components mixture model incorporates and functional covariates. This process can be thought as hierarchical with first level modelling according to parametric distributions second probabilities generalized linear The traditional approach treating covariates vectors not only suffers from curse dimensionality, since measured at very small intervals leading highly...
Tropical forage grasses, particularly those belonging to the Urochloa genus, play a crucial role in cattle production and serve as main food source for animals tropical subtropical regions. The majority of these species are apomictic tetraploid, highlighting significance U. ruziziensis , sexual diploid that can be tetraploidized use interspecific crosses with species. As means support breeding programs, our study investigates feasibility genome-wide family prediction families predict...
Abstract Dairy cows are responsible for a fair amount of gas emissions in the atmosphere (mainly methane, ammonia, and carbon dioxide), as well waste outputs. Therefore, identifying high‐fertility breeding increasing fertility rates can diminish pollution help minimize effect global warming improve environmental impact farming system. As step to achieve this goal, changes lipid composition bovine uterus exposed greater (LF‐LCL group) or lower (SF‐SCL concentrations progesterone during...
We consider unsupervised classification by means of a latent multinomial variable which categorizes scalar response into one L components mixture model. This process can be thought as hierarchical model with first level modelling according to parametric distributions, the second models probabilities generalised linear functional and covariates. The traditional approach treating covariates vectors not only suffers from curse dimensionality since measured at very small intervals leading highly...