Identifying Structural Phases of Policy Uncertainty: A Visibility Graph and Complex Network Approach to EPU and TPU

This study examines the phased characteristics of Economic Policy Uncertainty (EPU) and Trade Policy Uncertainty (TPU) utilizing visibility graph algorithms and complex network models. By employing a modularity optimization approach, a dynamic network structure of the time series was constructed, enabling precise delineation of the evolution patterns of uncertainty from 2001 to 2020 and the identification of distinct phases of policy fluctuations. Empirical analysis reveals that the model effectively captures the significant impacts of major events—such as the International Financial Crisis and the U.S.–China Trade War—on policy uncertainty, as well as the structural disturbance observed around the early stage of COVID-19 in 2020. In doing so, the model uncovers the dynamic properties of different periods and their external driving forces within the 2001–2020 sample window. The findings demonstrate that this methodology can adeptly unveil the temporal dynamics of uncertainty, providing robust support for policy formulation, risk assessment, and economic environment research.