Classification of Japanese Electrical Equipment Manufacturing Industry Recovery Patterns after Disasters: Case Study of August 2019

Abstract

In the paper, we analyze the recovery pattern of Japanese electrical equipment manufacturing companies after the President Trump remark in August 2019. The President’s remark made the companies’ stock prices decreased severely. The research consists of two parts. In the first part, we conducted Random Matrix Theory to extract representative decline/recovery patterns. Then we tagged A/B/C/D to the companies’ recovery types. The class A means a strong recover power. Then as the second part, we conducted machine learning tree-based classification using the tags A/B/C. The predictors are eight variables like ROA, ROE, and VAR. The resultant Decision Tree model provided us with the two different approaches to the class A group. The recovery and repulsion power will be higher in the company with high ROA and in the company that manufactured the product with high VAR. In addition, another class A company group is made and the feature is the high inventory turnover ratio.