Calibration of weights in network-based credit contagion models

Determine a robust and practical method to calibrate the edge weights in the directed weighted graph specifying inter-firm dependencies in network-based credit contagion models, specifically the framework of Egloff, Leippold, and Vanini (2007), to enable empirical implementation of such models.

Background

Within the literature review, the paper discusses models that incorporate micro-structural interdependencies via networks, notably Egloff et al. (2007), where firms are connected in a directed weighted graph. These approaches can capture contagion but require specifying weights that quantify the strength of inter-firm links.

The authors explicitly note that calibrating these network weights remains an unresolved issue not only for Egloff et al.'s model but broadly for network-based credit contagion frameworks, posing a barrier to practical adoption and empirical validation.