A Panel Data Model with Generalized Higher-Order Network Effects

Many data situations require the consideration of network effects among cross-sectional units observation. In this paper, we present a generalized panel model which accounts for two features: (i) three types on right-hand side model, namely through weighted dependent variable, exogenous variables, as well error components, and (ii) higher-order due to ex-ante unknown network-decay functions or presence multiplex (or multi-layer) networks all those. We outline basic assumptions, simulation results.