Rates of convergence and normal approximations for estimators of local dependence random graph models

Local dependence random graph models are a class of block models for network data which allow for dependence among edges under a local dependence assumption defined around the block structure of the network. Since being introduced by Schweinberger and Handcock (2015), research in the statistical network analysis and network science literatures have demonstrated the potential and utility of this class of models. In this work, we provide the first statistical disclaimers which provide conditions under which estimation and inference procedures can be expected to provide accurate and valid inferences. This is accomplished by deriving convergence rates of inference procedures for local dependence random graph models based on a single observation of the graph, allowing both the number of model parameters and the sizes of blocks to tend to infinity. First, we derive the first non-asymptotic bounds on the ₂-error of maximum likelihood estimators, along with convergence rates. Second, and more importantly, we derive the first non-asymptotic bounds on the error of the multivariate normal approximation. In so doing, we introduce the first principled approach to providing statistical disclaimers through quantifying the uncertainty about statistical conclusions based on data.