944 resultados para maximum likelihood method


Relevância:

80.00% 80.00%

Publicador:

Resumo:

Os objetivos neste trabalho foram estudar os efeitos de ambiente sobre a espessura de gordura subcutânea (EGS), a área de olho-de-lombo (AOL) e o peso aos 19 meses de idade e estimar parâmetros genéticos para essas características. Utilizaram-se informações obtidas de 987 bovinos da raça Canchim (5/8 Charolês + 3/8 Zebu) e do grupo genético animal MA (filhos de touros charoleses e vacas 1/2 Canchim + 1/2 Zebu) nascidos em 2003, 2004 e 2005. Os componentes de covariância foram estimados pelo método da máxima verossimilhança restrita utilizando-se um modelo animal com efeitos fixos (ano de nascimento, grupo genético, rebanho e sexo) e os efeitos aleatórios genético aditivo direto e residual. As médias de área de olho-de-lombo e peso foram mais altas nos machos que nas fêmeas. No grupo genético MA, as médias para todas as características foram mais altas que na raça Canchim e houve ainda efeitos de rebanho e de ano de nascimento. As estimativas de herdabilidade para AOL (0,33 ± 0,09), EGS (0,24 ± 0,09) e peso (0,23 ± 0,09) foram moderadas, enquanto que a estimativa de correlação genética (0,21 ± 0,24) entre EGS e AOL foi baixa, o que sugere que essas características são controladas por diferentes conjuntos de genes de ação aditiva. As correlações genéticas para peso estimadas com EGS (0,57 ± 0,23) e com AOL (0,62 ± 0,16) foram moderadas. Conclui-se que as características ao sobreano devem responder à seleção nos rebanhos estudados e que a seleção para aumento de peso também eleva EGS e AOL e vice-versa.

Relevância:

80.00% 80.00%

Publicador:

Resumo:

Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)

Relevância:

80.00% 80.00%

Publicador:

Resumo:

Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)

Relevância:

80.00% 80.00%

Publicador:

Resumo:

This work presents a study on the generation of digital masks aiming at edge detection with previously known directions. This solution is important when edge direction is available either from a direction histogram or from a prediction based on camera and object models. A modification in the non-maximum suppression method of thinning is also presented enabling the comparison of local maxima for any edge directions. Results with a synthetic image and with crops of a CBERS satellite images are presented showing an example with its application in road detection, provided that directions are previously known.

Relevância:

80.00% 80.00%

Publicador:

Resumo:

The work is to make a brief discussion of methods to estimate the parameters of the Generalized Pareto distribution (GPD). Being addressed the following techniques: Moments (moments), Maximum Likelihood (MLE), Biased Probability Weighted Moments (PWMB), Unbiased Probability Weighted Moments (PWMU), Mean Power Density Divergence (MDPD), Median (MED), Pickands (PICKANDS), Maximum Penalized Likelihood (MPLE), Maximum Goodness-of-fit (MGF) and the Maximum Entropy (POME) technique, the focus of this manuscript. By way of illustration adjustments were made for the Generalized Pareto distribution, for a sequence of earthquakes intraplacas which occurred in the city of João Câmara in the northeastern region of Brazil, which was monitored continuously for two years (1987 and 1988). It was found that the MLE and POME were the most efficient methods, giving them basically mean squared errors. Based on the threshold of 1.5 degrees was estimated the seismic risk for the city, and estimated the level of return to earthquakes of intensity 1.5°, 2.0°, 2.5°, 3.0° and the most intense earthquake never registered in the city, which occurred in November 1986 with magnitude of about 5.2º

Relevância:

80.00% 80.00%

Publicador:

Resumo:

This study includes the results of the analysis of areas susceptible to degradation by remote sensing in semi-arid region, which is a matter of concern and affects the whole population and the catalyst of this process occurs by the deforestation of the savanna and improper practices by the use of soil. The objective of this research is to use biophysical parameters of the MODIS / Terra and images TM/Landsat-5 to determine areas susceptible to degradation in semi-arid Paraiba. The study area is located in the central interior of Paraíba, in the sub-basin of the River Taperoá, with average annual rainfall below 400 mm and average annual temperature of 28 ° C. To draw up the map of vegetation were used TM/Landsat-5 images, specifically, the composition 5R4G3B colored, commonly used for mapping land use. This map was produced by unsupervised classification by maximum likelihood. The legend corresponds to the following targets: savanna vegetation sparse and dense, riparian vegetation and exposed soil. The biophysical parameters used in the MODIS were emissivity, albedo and vegetation index for NDVI (NDVI). The GIS computer programs used were Modis Reprojections Tools and System Information Processing Georeferenced (SPRING), which was set up and worked the bank of information from sensors MODIS and TM and ArcGIS software for making maps more customizable. Initially, we evaluated the behavior of the vegetation emissivity by adapting equation Bastiaanssen on NDVI for spatialize emissivity and observe changes during the year 2006. The albedo was used to view your percentage of increase in the periods December 2003 and 2004. The image sensor of Landsat TM were used for the month of December 2005, according to the availability of images and in periods of low emissivity. For these applications were made in language programs for GIS Algebraic Space (LEGAL), which is a routine programming SPRING, which allows you to perform various types of algebras of spatial data and maps. For the detection of areas susceptible to environmental degradation took into account the behavior of the emissivity of the savanna that showed seasonal coinciding with the rainy season, reaching a maximum emissivity in the months April to July and in the remaining months of a low emissivity . With the images of the albedo of December 2003 and 2004, it was verified the percentage increase, which allowed the generation of two distinct classes: areas with increased variation percentage of 1 to 11.6% and the percentage change in areas with less than 1 % albedo. It was then possible to generate the map of susceptibility to environmental degradation, with the intersection of the class of exposed soil with varying percentage of the albedo, resulting in classes susceptibility to environmental degradation

Relevância:

80.00% 80.00%

Publicador:

Resumo:

The information in this study has been provided by the Brazilian Association of Racehorse Breeders [Associação do Brasileira dos Criadores do Cavalo de Corrida (ABCCC)]. It can be found in the files on the CD-ROM developed by the ABCCC in 1999. A total of 5008 finishing time records related to 2545 winning horses that ran in the classical calendar on Brazilian hippodromes during 25 years (197498) were analysed. There were a total of 9949 horses on the relationship matrix. The variance components were estimated using the multiple-trait derivate-free restricted maximum likelihood (MTDFREML) program, for an animal model. Generation intervals were higher in the maternal side (10.91 years) than in the paternal one (10.41 years). The estimates for genetic permanent environmental and phenotypic variances and heritability were 0.291, 0.161, 3.486 and 0.08, respectively. The phenotypic standard deviation for time in races was 1.86729 s. Genetic time trend on Thoroughbred races in Brazil was small and could be accelerated if selection considered the trait time effectively. With respect to the animal's country of birth, the results show that there has been an intense participation of foreign animals in breeding Brazilian Thoroughbreds.

Relevância:

80.00% 80.00%

Publicador:

Resumo:

Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)

Relevância:

80.00% 80.00%

Publicador:

Resumo:

In this paper, we derive score test statistics to discriminate between proportional hazards and proportional odds models for grouped survival data. These models are embedded within a power family transformation in order to obtain the score tests. In simple cases, some small-sample results are obtained for the score statistics using Monte Carlo simulations. Score statistics have distributions well approximated by the chi-squared distribution. Real examples illustrate the proposed tests.

Relevância:

80.00% 80.00%

Publicador:

Resumo:

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

Relevância:

80.00% 80.00%

Publicador:

Resumo:

The family Callichthyidae comprises eight genera of fishes widely distributed across the Neotropical region. In the present study, sequences of the mitochondrial genes 12S rRNA, 16S rRNA, ND4, tRNA(His), and tRNA(Scr) were obtained from 28 callichthyid specimens. The sample included 12 species of Corydoras, three species of Aspidoras, two species of Brochis, Dianema, Lepthoplosternum, and Megalechis, and two local populations of Callichthys and Hoplosternum. Sequences of Nematogenys inermis (Nematogenyidae), Trichomycterus areolatus, and Henonemus punctatus (Trichomycteridae), Astroblepus sp. (Astroblepidae), and Neopleeostomus paranensis, Delturus parahybae, and Hemipsilichthys nimius (Loricariidae) were included as the outgroup. Phylogenetic analyses were performed by using the methods of maximum parsimony and maximum likelihood. The results of almost all analyses were very similar. The family Callichthyidae is monophyletic and comprises two natural groups: the subfamilies Corydoradinae (Aspidoras, Brochis, and Corydoras) and Callichthyinae (Callichthys, Dianema, Hoplosternum, Lepthoplosternum, and Megalechis), as previously demonstrated by morphological studies. The relationships observed within these subfamilies are in several ways different from those previously proposed on the basis of morphological data. Molecular results were compared with the morphologic and cytogenetic data available on the family. (C) 2003 Elsevier B.V. All rights reserved.

Relevância:

80.00% 80.00%

Publicador:

Resumo:

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

Relevância:

80.00% 80.00%

Publicador:

Resumo:

Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)

Relevância:

80.00% 80.00%

Publicador:

Resumo:

Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)

Relevância:

80.00% 80.00%

Publicador:

Resumo:

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