A0287
Title: Two-way homogeneity pursuit for quantile network vector autoregression
Authors: Xuening Zhu - Fudan University (China)
Ganggang Xu - University of Miami (United States) [presenting]
Wenyang Liu - Fudan University (China)
Jianqing Fan - Princeton University (United States)
Abstract: While the vector autoregression (VAR) model has received extensive attention for modeling complex time series, quantile VAR analysis remains relatively underexplored for high-dimensional time series data. To address this disparity, a two-way grouped network quantile (TGNQ) autoregression model is introduced for time series collected on large-scale networks, known for their significant heterogeneous and directional interactions among nodes. The proposed model simultaneously conducts node clustering and model estimation to strike a balance between complexity and interpretability. To account for the directional influence among network nodes, each network node is assigned two latent group memberships that can be consistently estimated using the proposed estimation procedure. The approach extends the homogeneity pursuit introduced in another study for VAR models, offering attractive asymptotic properties. Theoretical analysis demonstrates the consistency of membership and parameter estimators even with an over-specified number of groups. With the correct group specification, estimated parameters are proven to be asymptotically normal, enabling valid statistical inferences. Moreover, a quantile information criterion is proposed for consistently selecting the number of groups. Simulation studies show promising finite sample performance, and the methodology is applied to analyze connectedness and risk spillover effects among Chinese A-share stocks.
