EcoSta 2023: Start Registration
View Submission - EcoSta2023
A1332
Title: Oil shock, policy uncertainty and stock return: An analysis based on the Bart method Authors:  Jianhua Zhou - Sun Yat-Sen university (China) [presenting]
Abstract: An extension of the time-varying network dependence panel (TNDP) mode that incorporates the identification of dominant units is utilized. The objective is to re-examine the impacts of the oil shock and policy uncertainty on stock returns in a network framework with heterogeneity both over time and across sections. Bayesian econometric methods using additive regression trees are employed and are used to address extreme observations. We argue that regression tree models are well-suited for macroeconomic nowcasting due to their flexibility and ability to model outliers.