ISBA World Meeting 2024
Poster apresentado pela pesquisada Lilia Carolina Carneiro da Costa
Neste trabalho, Lilia apresentou um pacote baseado na linguagem R, capaz de ajustar modelos de redes bayesianas dinâmicas. Em particular, se tem interesse em estudar a relação causal de series temporais, como por exemplo, a transmissão de uma doença entre as diferentes localidades brasileiras.
A Multiregression Dynamic Model (MDM) is a Bayesian class of multivariate time series that represents various dynamic causal processes in a graphical way. Because the marginal likelihood has a closed form, model selection across many potential connectivity networks is easy to perform. With the application of the Integer Programming Algorithm, we can quickly find optimal models that satisfy acyclic graph constraints, and due to a factorization of the marginal likelihood, the search over all possible directed (acyclic or cyclic) graphical structures is even faster. The package mdmr implements the method developed by Costa et al. (2015) in R language, open source, and can be accessed at https://github.com/arzevedo/mdmr. In this work, we will present two applications that show the method’s versatility using this package. In the first, we will investigate the relationship between new cases of COVID-19 in Brazilian macro-regions, and in the second one, we will use it to estimate inter-brain connectivity.
Mais informações podem ser encontradas no website oficial do evento