Generalized linear mixed models: an application in fungal occurrence data

Thiago de Paula Oliveira, Rafael A. Moral, John Hinde, Clarice G.B. Demetrio, Silvio Sandoval Zocchi

Research output: Contribution to conferenceAbstract

Abstract / Description of output

When analysing proportion data, a useful framework is that of generalized linear models. Random effects may be included in the linear predictor
for different reasons, e.g., to incorporate correlation between observations taken
within the same subject or to model overdispersion. In this work, we use binomial
mixed models to model the occurrence of entomopathogenic fungi in five different
Brazilian biomes in the dry and humid seasons of 2012. We add an observationlevel random effect to incorporate overdispersion and test for the significance of
the interaction effect between biome and season
Original languageEnglish
Publication statusPublished - 8 Jun 2015
Event60 Reunião Anual da Região Brasileira da Sociedade Internacional de Biometria e 16 Simpósio de Estatística Aplicada à Experimentação Agronômica - Presidente Prudente, Brazil
Duration: 8 Jun 201514 Jun 2015

Conference

Conference60 Reunião Anual da Região Brasileira da Sociedade Internacional de Biometria e 16 Simpósio de Estatística Aplicada à Experimentação Agronômica
Country/TerritoryBrazil
CityPresidente Prudente
Period8/06/1514/06/15

Fingerprint

Dive into the research topics of 'Generalized linear mixed models: an application in fungal occurrence data'. Together they form a unique fingerprint.

Cite this