Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
We consider the aggregation of heterogeneous dynamic equations across a large population, as introduced by Granger (1980), where the dynamics arise because agents face a signal extraction problem caused by incomplete information. This weakens the independence assumptions used previously in the aggregation literature. We show that, under plausible assumptions, the differenced cross-section aggregate shows long term persistence even though every individual micro-series follows a random walk. As an example, estimates of the model’s micro-relations are made using US household panel data.