Vector Autoregression

Vector autoregression (VAR) models analyze the interdependencies between multiple time series, aiming to understand and predict their evolution. Current research focuses on improving VAR's scalability for large datasets, incorporating prior knowledge (e.g., physical laws) into model design, and developing methods for causal inference and counterfactual analysis within the VAR framework. These advancements are enhancing the applicability of VAR models across diverse fields, including finance, healthcare, and climate science, by enabling more accurate predictions and a deeper understanding of complex dynamical systems.