Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
This paper introduces Mixed Data Sampling (MIDAS) into the panel data context. To address the unidentified nuisance parameter problem, we propose to invert model specification tests for inference on the MIDAS parameter along with bounds tests for model coefficients. Illustrative identification, simulation and empirical analyses are conducted in the dynamic GMM framework. Our framework allows for departures from i.i.d errors such as clustering and dynamic specifications. A simulation study and an application to a model of reserve holdings illustrate the usefulness of the proposed methods, and more broadly set a promising template for shrinkage approaches.