A0471
Title: Network varying coefficient model
Authors: Xinyan Fan - Renmin University of China (China) [presenting]
Wei Lan - Southwestern University of Finance and Economics (China)
Kuangnan Fang - Xiamen University (China)
Abstract: A novel network varying coefficient model (NVCM) is proposed that extends traditional varying coefficient models (VCM) to accommodate network data. The key idea is to model the regression coefficients as functions of the latent locations of network nodes that drive the formation of the network. To estimate the model, the latent locations are identified via the latent space model, and an iterative projected gradient descent algorithm is developed by maximizing the network parameters and regression coefficients alternately. 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. The method is further combined with a penalization procedure to select covariates with varying coefficients, as well as those that are significant to the response variable 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.
