A Dynamic Network Regression Model for a Large Cross Section of Units With an Application to Measuring Spillovers Between Pollution and Electricity Consumption

This paper proposes a dynamic regression model that incorporates network effects from a large cross section of units. We consider a functional coefficient to measure these effects that is estimated using Taylor expansions over a partition of disjoint intervals of the compact support of the network variable. This method yields consistent and asymptotically normal estimates of the functional coefficient. We also derive uniform convergence results for the nonparametric estimator of the functional process capturing network effects. The model is applied to estimate the relationship between electricity consumption and carbon (NO2) emissions for a set of European countries for the years 2019 and 2020, separately. We find evidence of a quadratic relationship between both variables and uncover spatial effects from neighboring countries