A Measurement of Financial Network Connectedness
演讲者
时间
2024年10月28日 15:20 至 16:20
地点
A3-2-303
线上
Zoom 230 432 7880
(BIMSA)
摘要
This talk introduces one of the most popular network methodologies currently used in economics and finance , FX Diebold, K Yilmaz(2009) proposed this method subsequently refined through a series of papers (DY2012, DY2014, DY2018), which is constructed based on forecast-error variance decompositions (FEVD) from generalized VAR models. This approach can intuitively characterize both static and dynamic connectedness relationships among different asset classes and companies across different countries. In DY2014, the authors recognized that their methodology belongs to the broader field of network science, with their spillover effect matrix being equivalent to an adjacency matrix. DY2018 further enhanced the framework by incorporating LASSO, elastic net, and other methods to shrink, select, and estimate high-dimensional networks. Building upon these works, Gabauer and Gupta (2018) extended the methodology by replacing the GVAR model with a TVP-VAR model, thereby eliminating the need for rolling windows. This talk also implements some sample applications using the R package 'ConnectednessApproach'.
演讲者介绍
Junda Wu is a PhD student at BIMSA and UCAS. His research interests include digital economy, applications of artificial intelligence in economics, and cross-market risk spillovers.