Autoregressive Network: Sparsity and Degree Heterogeneity

We studied a type of dynamic network model where edges between each pair of nodes change over time independently. In this model, the connecting probability and disconnecting probability are allowed to go to $0$ as the number of nodes goes to infinity. We proposed a two-step estimation strategy, where the first step is MLE and the second step is an M-estimation followed by the uniqueness of the solution.