Team Networks with Partially Observed Links

TL;DR

This paper introduces a new estimator for team network models with missing links, addressing partial observability and bias correction, with applications to academic collaboration data.

Contribution

It develops a Generalized Method of Moments estimator and a distribution-free test for link truncation in partially observed team networks.

Findings

Reveals downward bias in estimated scaling factors due to missing links.

Corrects bias in estimating team-specific fixed effects.

Suggests collaboration premiums are underestimated without accounting for missing links.

Abstract

This paper studies a linear production model in team networks with missing links. In the model, heterogeneous workers, represented as nodes, produce jointly and repeatedly within teams, represented as links. Links are omitted when their associated outcome variables fall below a threshold, resulting in partial observability of the network. To address this, I propose a Generalized Method of Moments estimator under normally distributed errors and develop a distribution-free test for detecting link truncation. Applied to academic publication data, the estimator reveals and corrects a substantial downward bias in the estimated scaling factor that aggregates individual fixed effects into team-specific fixed effects. This finding suggests that the collaboration premium may be systematically underestimated when missing links are not properly accounted for.