主讲人：杨子欣 助理研究员（上海财经大学统计与数据科学学院）

主持人：兰兰 助理教授（辽宁大学李安民经济研究院）

嘉宾介绍：毛明海 助理教授（辽宁大学李安民经济研究院）

时间：2025年11月13日（周四） 10：00-11：30（北京时间）

地点：辽宁大学崇山校区五洲园一楼会议室

线上地址：腾讯会议730-281-8924

语言：中文/英文

摘要： This paper examines the nonlinear dynamics of a high-dimensional time series collected from the nodes of a large-scale network. We propose a first-order vector autoregressive model that incorporates nonlinear and nonparametric network and momentum effects. The model further accommodates node-level heterogeneity in the response functions. For estimation, we develop a sieve least squares estimator and establish its consistency and asymptotic normality under various asymptotic regimes. Furthermore, a specification test is proposed to examine the linearity of the nonparametric components. The test statistic has an asymptotically standard normal distribution under the null hypothesis and a sequence of Pitman local alternatives. The usefulness of the proposed model is illustrated through a user activity analysis on Sina Weibo, China's largest Twitter-like social media platform.

主讲人简介：

杨子欣，上海财经大学统计与数据科学学院助理研究员，主要研究领域为空间计量经济学、网络计量经济学、非参数和半参数方法等。研究论文发表在Journal of Econometrics, Journal of Business & Economic Statistics等学术期刊上。