报告人：于鑫洋（伦敦政治经济学院）

报告时间：2026年1月14日（周三）上午 9:30-12:00 

报告地点：国交2号楼315会议室

报告摘要：Statistical modeling of network data is an important topic in various areas. Al-

though many real networks are dynamic in nature, most existing statistical models

and related inferences for network data are confined to static networks, and the de-

velopment of the foundation for dynamic network models is still in its infancy. In

particular, to the best of our knowledge, no attempts have been made to jointly ad-

dress node heterogeneity and link homophily among dynamic networks. Being able

to capture these network features simultaneously will not only bring new insights

on understanding how networks were formed, but also provide more sophisticated

tools for the prediction of a future network with statistical guarantees. In particu-

lar, our model accounts for link homophily associated with both observed traits and

latent traits of the nodes. A novel normalized least squared loss based framework

is constructed to generate stable estimations for the high dimensional parameters.

The promising performance of the proposed model is further illustrated by various

simulation and real data studies.