A0940
Title: A two-way heterogeneity model for dynamic networks
Authors: Binyan Jiang - The Hong Kong Polytechnic University (Hong Kong) [presenting]
Abstract: Analysis of networks that evolve dynamically requires the joint modelling of individual snapshots and time dynamics. The aim is to propose a new flexible two-way heterogeneity model towards this goal. The new model equips each node of the network with two heterogeneity parameters, one to characterize the propensity to form ties with other nodes statically and the other to differentiate the tendency to retain existing ties over time. With n observed networks, each having p nodes, a new asymptotic theory for the maximum likelihood estimation of $2p$ parameters is developed when $np\rightarrow\infty$. The global non-convexity of the negative log-likelihood function is overcome by virtue of its local convexity, and a novel method of moment estimator is proposed as the initial value for a simple algorithm that leads to the consistent local maximum likelihood estimator (MLE). To establish the upper bounds for the estimation error of the MLE, a new uniform deviation bound is derived, which is of independent interest. The theory of the model and its usefulness are further supported by extensive simulation and real data analysis.
