A statistical process control method for monitoring social networks using generalized likelihood ratio test

摘要

Statistical modeling and analysis of social networks has evolved into an important area of research related to the complex relationships among social entities. In this paper, a new monitoring method based on Exponential Random Graph Model (ERGM) is presented and control charts are used to detect out-of-control states in networks. The Markov Chain Monte Carlo (MCMC) algorithm is used for parameter estimation purposes. A monitoring statistics based on the Generalized Likelihood Ratio Test (GLRT) is applied to detect departures to anomalies in networks. A case study in social network is used to depict the properties and benefits of the proposed methodology, and the results show that GLR chart has better performance in average run length (ARL) as compared to the CUSUM chart in detecting large shifts, while the CUSUM chart is better for detecting small shifts.