Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
We propose a new semiparametric autoregressive duration (SACD) model, which incorporates the parametric and nonparametric estimators of the conditional duration in a multiplicative way. Asymptotic properties for this combined estimator are presented. The empirical application to the transaction duration of the US 2-Year Treasury note shows the outperformance of our SACD models over parametric ACD models.