Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
We introduce weighted links in a local public good game in an endogenous network with heterogeneous players. We find that the equilibrium predictions are sharper than when links are not weighted. In particular, active players form a complete core-periphery graph, where they are either in the core of interconnected players, or connected to every player in the core. Furthermore, a player's type is tightly related to her public good provision and her position in the network.