Distribution Vector Autoregression: Eliciting Macro and Financial Dependence

Speaker: Associate Professor Dan Zhu
Affiliation: Monash University
Location: Online via Zoom
Zoom link: https://uqz.zoom.us/j/85606810857 

Abstract 

The global financial crisis prompted an increased focus on the role of financial factors in driving real economic fluctuations. Vector autoregression models are popular in this literature, providing simple yet insightful information such as the impulse response function of different shocks. This paper develops a flexible and robust alternative based on a multivariate Distribution Regression method. The resulting distribution impulse response function provides a more comprehensive picture of the dynamic heterogeneity. In the study of the U.S. GDP growth and financial conditions, the empirical results from our new framework confirm some existing findings in the literature that 1) the tight financial condition creates multimodality in the conditional joint distribution and 2) restricting the upper tail of financial condition has a noticeable impact on long-term GDP growth. Yet, the extracted information on the effect of restricting the lower tail of GDP during the global financial crisis suggests an alternative conclusion, i.e., negligible impact on financial condition.

About the presenters meeting

A/Prof Zhu is happy to have (30-min) 1-to-1 meeting with our colleagues and graduate students on the seminar day. Anyone who wants to meet with her can contact Dr Fu Ouyang to book a time slot.