An analysis of political turmoil effects on stock prices: a case study of US-China trade friction

In the paper, we report an interesting result of changes of stock prices due to a political turmoil, the trade friction between China and US ignited in 2018, using the machine learning approach based on hierarchical clustering and Singular Value Decomposition methods and show such new approaches' possibilities and meaningfulness. The data we used are the top 100 global automobile manufactures' stock prices from 2018 to 2019 which were under the trade friction turmoil. The involved countries are Germany, Japan and US. One clear result is that the turmoil gave distinctively different effects on those countries' stock markets. We could identify three different clusters of stock price movements, that is, German, Japanese and US clusters. This result is expected to give some insights to the issue of international linkages between the movements of the markets' prices by adding a case of political turmoil.