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
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 language | English |
---|---|
Publication status | Published - 8 Jun 2015 |
Event | 60 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 2015 → 14 Jun 2015 |
Conference
Conference | 60 Reunião Anual da Região Brasileira da Sociedade Internacional de Biometria e 16 Simpósio de Estatística Aplicada à Experimentação Agronômica |
---|---|
Country/Territory | Brazil |
City | Presidente Prudente |
Period | 8/06/15 → 14/06/15 |