报告主题：Network Varying Coefficient Model
We propose in this article a novel network varying coefficient model (NVCM) that extends traditional varying coefficient models (VCM) to accommodate network data. Specifically, we model the regression coefficients of network nodes as functions of their latent locations within the latent space, where the latent locations are identified via the notable latent space model of Hoff et al. (2002). To estimate the model with unknown latent locations variables, we develop an iterative projected gradient descent algorithm. The non-asymptotic bounds of the estimated coefficients matrix are obtained theoretically. Practically, the dimension of the latent space is chosen via a Bayesian information criterion (BIC)- type criterion. We further combine our method with a penalization procedure to choose relevant latent variables and derive the related theoretical properties. The utility of the model is further illustrated via simulation studies as well as a real-world application in the field of finance by analyzing the relationship between stock returns and firm characteristics from a network perspective. The results show that the proposed model outperforms most existing methods.
Joint work with Wei Lan and Kuangnan Fang
范新妍，2019年厦门大学取得博士学位，现任中国人民大学统计学院副教授。主要研究方向多源数据分析，网络数据分析。研究工作发表在AoAS, JBES, Sinica 等期刊。