Measuring Connectedness in High-Dimensional Networks: Lasso-ing Variance Decomposition

Themabeschreibung

use Lasso to shrink and estimate sparse VAR
derive sparse variance decomposition matrix to quantify connectedness clearly
Empirical study: data mostly ready to use
Economic, Statistical techniques used: LASSO for VAR, variance decomposition
references:
- Demirer et al. (2015): Estimating global bank network connectedness
- Tibshirani (1996): Regression shrinkage and selection via the lasso
- Lütkepohl (2005): New introduction to multiple time series analysis