Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
Spillover indices computed from VAR models are intended to measure the connectedness between the variables in the system. The generalized spillover index (gSOI) computed using the generalized forecast error variance decomposition is often considerably larger than the conventional spillover index computed from specific Cholesky decompositions leading to the speculation that the gSOI produces an unreasonable measure of connectedness. We demonstrate that the gSOI does not produce unrealistic values.