Modelling longitudinal polytomous animal data using Bayesian hierarchical models

Resumo

e analysis of longitudinal categorical data can be complex and unfeasible due to the number of parameters involved, characterised by overparameterisation leading to model non-convergence, in addition to problems related to sample size and the presence or absence of overdispersion. In this context, we adopt the Bayesian hierarchical models as an alternative methodology to classical statistical techniques for analysing nominal polytomous data in longitudinal studies. The theoretical foundation is based on the use of non-informative priors and parameter estimation performed via Markov Chain Monte Carlo (MCMC) methods, which enable a robust and flexible data analysis framework. As a motivating example, the procedure is illustrated through an applied study in agrarian science, focusing on animal welfare, which assessed seven types of behaviours exhibited by pigs over twelve weeks. Model comparison based on the deviance information criterion (DIC) indicated that the model including the sex and enrichment interaction provided the best fit to the data. The results illustrate the applicability of Bayesian hierarchical models for the analysis of longitudinal nominal polytomous data. Since the computational procedures were implemented in the R software and the codes are available, this work will serve as support for those who need such analyses, especially in agricultural designs, where longitudinal categorical data are frequently encountered.The analysis of longitudinal categorical data can be complex and unfeasible due to the number of parameters involved, characterised by overparameterisation leading to model non-convergence, in addition to problems related to sample size and the presence or absence of overdispersion. In this context, we adopt the Bayesian hierarchical models as an alternative methodology to classical statistical techniques for analysing nominal polytomous data in longitudinal studies. The theoretical foundation is based on the use of non-informative priors and parameter estimation performed via Markov Chain Monte Carlo (MCMC) methods, which enable a robust and flexible data analysis framework. As a motivating example, the procedure is illustrated through an applied study in agrarian science, focusing on animal welfare, which assessed seven types of behaviours exhibited by pigs over twelve weeks. Model comparison based on the deviance information criterion (DIC) indicated that the model including the sex and enrichment interaction provided the best fit to the data. The results illustrate the applicability of Bayesian hierarchical models for the analysis of longitudinal nominal polytomous data. Since the computational procedures were implemented in the R software and the codes are available, this work will serve as support for those who need such analyses, especially in agricultural designs, where longitudinal categorical data are frequently encounteredseThe analysis of longitudinal categorical data can be complex and unfeasible due to the number of parameters involved, characterised by overparameterisation leading to model non-convergence, in addition to problems related to sample size and the presence or absence of overdispersion. In this context, we adopt the Bayesian hierarchical models as an alternative methodology to classical statistical techniques for analysing nominal polytomous data in longitudinal studies. The theoretical foundation is based on the use of non-informative priors and parameter estimation performed via Markov Chain Monte Carlo (MCMC) methods, which enable a robust and flexible data analysis framework. As a motivating example, the procedure is illustrated through an applied study in agrarian science, focusing on animal welfare, which assessed seven types of behaviours exhibited by pigs over twelve weeks. Model comparison based on the deviance information criterion (DIC) indicated that the model including the sex and enrichment interaction provided the best fit to the data. The results illustrate the applicability of Bayesian hierarchical models for the analysis of longitudinal nominal polytomous data. Since the computational procedures were implemented in the R software and the codes are available, this work will serve as support for those who need such analyses, especially in agricultural designs, where longitudinal categorical data are frequently encountered. de dados categóricos longitudinais pode ser complexa e inviável devido ao número de parâmetros envolvidos, caracterizando-se por uma superparametrização que leva à não convergência do modelo, além de problemas relacionados ao tamanho da amostra e à presença ou ausência de superdispersão. Nesse contexto, apresentamos os modelos hierárquicos bayesianos como uma metodologia uma alternativa às técnicas estatísticas clássicas para a análise de dados nominais políticos em estudos longitudinais. A fundamentação teórica baseia-se no uso de anteriores não informativos e em técnicas computacionais avançadas, como os métodos de Cadeia de Markov Monte Carlo (MCMC), que permitem um quadro de análise de dados robusto e flexível. Como exemplo motivador, o procedimento é ilustrado por meio de um estudo aplicado em ciências agrárias, com foco no bem-estar animal, que avaliou sete tipos de comportamentos exibidos por suínos ao longo de doze semanas. Os resultados demonstraram a eficácia dos modelos hierárquicos Bayesianos na análise de dados nominais politômicos longitudinais. Como os procedimentos computacionais foram implementados no software R e os códigos estão disponíveis, este trabalho serve como suporte para aqueles que ocorrem tais análises, especialmente em experimentos agrícolas, cujos dados categóricos longitudinais são frequentemente encontrados.