Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
In this note we extend the method proposed in Bun and Carree (2006) to the more general PVARX(1) model and show that the iterative procedure is not consistent for fixed T. Subsequently we provide corrected version of the bias correction procedure which is fixed T consistent and robust to both cross-sectional and time-series heteroscedasticity.