• JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
 
  Bookmark and Share
 
 
Master's Dissertation
DOI
https://doi.org/10.11606/D.11.2022.tde-13092022-094518
Document
Author
Full name
Antonio Leopoldo Cardoso Sabino
E-mail
Institute/School/College
Knowledge Area
Date of Defense
Published
Piracicaba, 2022
Supervisor
Committee
Zocchi, Silvio Sandoval (President)
Ferreira, Iuri Emmanuel de Paula
Leandro, Roseli Aparecida
Title in Portuguese
Modelo beta misto bayesiano para descrever a influência de diferentes cepas do fungo Colletotrichum truncatum sobre a severidade da antracnose na sojicultora
Keywords in Portuguese
MCMC
Patógeno
Soja
Taxas
Abstract in Portuguese
A cultura da soja é muito importante na economia mundial, pois sua matéria prima é utilizada para produção de diversos produtos. As plantas de soja, entretanto, são suscetíveis a diversas doenças causadas por fungos, como a antracnose, que pode chegar a comprometer a safra inteira. Essa doença é causada por cepas do fungo Colletotrichum truncatum, pertencentes a grupos genéticos distintos, que podem levar a índices de severidade diversos. Nas análises estatísticas usuais de dados de severidade, entretanto, frequentemente considera-se, equivocadamente, que seguem uma distribuição normal ou são realizadas, previamente, transformações dos dados, como logit, probit ou complemento log-log, dentre outras. Como a severidade é uma variável contínua entre 0 e 100%, (ou entre 0 e 1), a distribuição beta pode ser mais apropriada e metodologias como a proposta por Cribari-Neto e Zeileis (2010) podem ser consideradas. Em experimentos de comparação de grupos genéticos de cepas de antracnose quando ao índice de severidade podemos considerar o efeito de cepas como fixo ou aleatório dependendo do interesse do pesquisador. Apresentamos, aqui, uma abordagem bayesiana para a análise de dados provenientes de experimentos inteiramente ao acaso do tipo, com dois fatores, grupo genético e cepas. A metodologia foi implementada utilizando-se a interface RStan (Stan Development Team, 2020) e ilustrada por meio de um conjunto de dados reais e inéditos. A análise dos mesmos revela que diferentes conclusões podem ser obtidas ao considerarmos efeitos de cepa fixos ou aleatórios.
Title in English
Bayesian mixed beta model to describe the influence of different strains of the fungus Colletotrichum truncatum on the severity of anthracnose in soybeans
Keywords in English
MCMC
Pathogen
Rates
Soybean
Abstract in English
Soybean farming is very important in the world economy, as soy is used for the production of various products. Soybean plants, however, are susceptible to several diseases caused by fungi, such as anthracnose, which can compromise the entire crop. This disease is caused by strains of the fungus Colletotrichum truncatum, belonging to different genetic groups, which can lead to different levels of severity. In the usual statistical analysis of severity data, however, it is often mistakenly considered that they follow a normal distribution or that data transformations are carried out in advance, such as logit, probit or log-log complement, among others. As the severity is a continuous variable between 0 and 100%, (or between 0 and 1), the beta distribution can be more appropriate and methodologies like the one proposed by Cribari-Neto e Zeileis (2010) can be considered. In experiments comparing genetic groups of anthracnose strains regarding the severity index, we can consider the effect of strains as fixed or random depending on the researcher’s interest. We present here a Bayesian approach to the analysis of data from entirely randomized experiments of this type, with two factors, genetic group and strain. The methodology was then implemented using the RStan interface (Stan Development Team, 2020) and illustrated using a set of unpublished real data. Their analysis reveals that different conclusions can be reached when considering fixed or random strain effects.
 
WARNING - Viewing this document is conditioned on your acceptance of the following terms of use:
This document is only for private use for research and teaching activities. Reproduction for commercial use is forbidden. This rights cover the whole data about this document as well as its contents. Any uses or copies of this document in whole or in part must include the author's name.
Publishing Date
2022-09-14
 
WARNING: Learn what derived works are clicking here.
All rights of the thesis/dissertation are from the authors
CeTI-SC/STI
Digital Library of Theses and Dissertations of USP. Copyright © 2001-2024. All rights reserved.