Research on Influencing Factors of Sino-US Trade Based on Principal Component Analysis
DOI:
https://doi.org/10.56028/aemr.14.1.646.2025Keywords:
Principal Component Analysis; Influencing Factors; Pearson Correlation Coefficient; Sino-US Trade; Trade Friction.Abstract
This paper focuses on Sino-US trade relations and explores the evolution of key influencing factors in different time periods. A 16-item three-level index evaluation system covering four dimensions—economic, political, technological, and social—was constructed. First, the Pearson correlation coefficient was used to preliminarily screen index data from 2005 to 2023, eliminating indicators with low correlation to Sino-US trade volume. The research period was then divided into two phases using principal component analysis (PCA): 2005–2017 and 2018–2023. Dimensionality reduction was performed for each phase, extracting 1 and 2 key principal components sequentially. The results show that economic and technological factors were the core drivers of Sino-US trade growth during 2005–2017. In the 2018–2023 phase, influenced by intensified trade friction and geopolitics, core drivers shifted significantly: the explanatory power of economic and technological foundations weakened, and the influence of many previously key indicators dropped sharply or became invalid. Political game factors emerged as an independent and important second principal component for the first time, indicating a significant increase in political intervention in Sino-US trade. Based on this, four suggestions are proposed: optimizing export structure to enhance industrial resilience, strengthening technological innovation to consolidate the technical foundation, improving policy mechanisms to optimize the business environment, and expanding diversified markets to reduce external risks. These aim to provide references for addressing the complex landscape of Sino-US trade.