Measuring Systemic Risk with Network Connectivity: Extended Abstract

Measuring connectedness among financial institutions is an important aspect of monitoring systemwide risk development and identifying systemically risky institutions. In this work, we present a novel statistical method for measuring connectivity among firms using publicly available time series of firm-level characteristics. The proposed method relies on a Lasso penalized estimation of high-dimensional vector autoregressive models (VAR) and provides a principled framework for estimating network topology of linkages among firms. We apply our method to analyze connectivity among stock returns of leading financial firms in the U.S. before, during and after the financial crisis of 2007-2009, and show that centrality measures of the estimated networks can be used to identify important systemic events and systemically risky institutions.