Network effects error components models

MONTES ROJAS, GABRIEL VICTORIO

CABA

Conferencia; IV JORNADAS ARGENTINAS DE ECONOMETRÍA; 2018

This paper develops a random effects error components structurefor network data regression models. In particular, it allows for edgeand triangle specific components, which serve as a basal model formodeling network effects. It then evaluates the potential effects ofignoring network effects in the estimation of the variance-covariancematrix. Network effects will typically imply heteroskedasticity, andas with the Moulton factor, the key role is given by the joint consideration of the intra-network correlation of the error term(s) and thecovariates. Then it proposes consistent estimator of the variance components and Lagrange Multiplier tests for evaluating the appropriatemodel of random components in networks. Monte Carlo simulationsshow the tests have very good performance in finite samples.