Statistical inference for noisy dynamic networks

Music City Center 

In this work we develop techniques to estimate and perform inference on subgraph densities using time-indexed network sequences. These estimates explicitly adjust for observation errors for the network edges, and have good theoretical properties as the size of the network grows. By specifying a stochastically evolving hidden Markov network model, we are able to address two areas left for further investigation by Chang et al. (2022): robustness to non-identical network replicates, and efficient aggregation of all available data. Adaptation to these new settings vastly expands the applicability of their methods to real data settings, where network replicates are commonly observed dynamically. The methodology is also extended to consider joint inference for subgraph densities at multiple time points, to facilitate formal statistical comparison of dynamic network snapshots.

Dynamic network

Network inference

Hidden Markov model

Observation error 

Section on Statistical Learning and Data Science