Abstract

We propose a network model for expected shortfall (ES) to identify the network spill-
over effects between firms' tail risk exposures. The proposed model is an extension of
vector autoregressive regression (VAR) to tail based observations.  We derive
Generalized Method of Moments-based estimators and establish the consistency and
asymptotic normality of the estimator for model parameters. Based on the model
estimation, we further construct a series of measures to identify the roles of firms
in terms of network risk transmission within the system, and examines the pricing of
this "network risk" in the cross section of firms' stock returns.