Miniconferência 1 – 31/05 (10:30 – 11:00) - Anfiteatro

Palestrante: Estevão Prado – Lancaster University, UK

We propose some extensions to semi-parametric models based on Bayesian additive regression trees (BART). In the semi-parametric BART paradigm, the response variable is approximated by a linear predictor and a BART model, where the linear component is responsible for estimating the main effects and BART accounts for non-specified interactions and non-linearities. Previous semi-parametric models based on BART have assumed that the set of covariates in the linear predictor and the BART model are mutually exclusive in an attempt to avoid poor coverage properties and reduce bias in the estimates of the parameters in the linear predictor. The main novelty in our approach lies in the way we change the tree-generation moves in BART to deal with this bias and resolve non-identifiability issues between the parametric and non-parametric components, even when they have covariates in common. This allows us to model complex interactions involving the covariates of primary interest, both among themselves and with those in the BART component. Our novel method is developed with a view to analyzing data from an international education assessment, where certain predictors of students’ achievements in mathematics are of interpretational interest. Through additional simulation studies and another application to a well-known benchmark dataset, we also show competitive performance when compared to regression models, alternative formulations of semi-parametric BART, and other tree-based methods. The implementation of the proposed method is available at

Miniconferência 2 – 31/05 (11:00 – 11:30) - Anfiteatro

Palestrante: Adèle Helena Ribeiro – Philipps University of Marburg and Heinrich Heine University of Düsseldorf, Germany

Within the realm of empirical sciences, a common challenge arises when attempting to ascertain causal relationships from non-experimental, observational data. The process of inferring causality in observational studies often requires a deep understanding of the underlying causal structure, typically represented as a causal graph. Unfortunately, in practice, this essential structural knowledge is often scarce or absent, thereby impeding the establishment of robust and precise causal estimations. In my talk, I will discuss a few recent advancements in the field that offer valuable tools for researchers and data scientists seeking to uncover causal effects in real-world scenarios where prior knowledge is largely coarse and imprecise. Specifically, I will delve into recently proposed techniques in the areas of causal modeling, causal discovery, and causal effect identification that aim to bridge the gap between theoretical concepts and their practical implementation.