Estimation of a Scale-Free Network Formation Model

Growing evidence suggests that many social and economic networks are scale free in that their degree distribution has a power-law tail. A common explanation for this phenomenon is a random network formation process with preferential attachment. For a general version of such a process, we develop the pseudo maximum likelihood and generalized method of moments estimators. We prove consistency of these estimators by establishing the law of large numbers for growing networks. Simulations suggest that these estimators are asymptotically normally distributed and outperform the commonly used non-linear least squares and Hill (1975) estimators in finite samples. We apply our estimation methodology to a co-authorship network