Host: The Japanese Society for Artificial Intelligence
Name : 34th Annual Conference, 2020
Number : 34
Location : Online
Date : June 09, 2020 - June 12, 2020
In order to confirm propagation of economic shock through supply chains, we create a regression model that expresses the relationship between supplier's sales growth ratio and that of their customers. From a database having past financial information of listed companies, we get firms' net sales from 2012 to 2016 (approximately 7,500 cases), regardless of they were affected by accidents or disasters. By collating this information with another database having business relationships of firms, we create a data set containing sales growth ratio of suppliers and that of their customers in each fiscal year. Then we estimate customers' sales growth ratio by using a machine learning (LightGBM) with the suppliers' sales growth ratio and someother information. As a result, we have confirmed that we can estimate it with acceptable accuracy by using suppliers' sales growth ratio as one of explanatory variables. In other words, we have shown that firms' sales growth is likely to be affected by the performance of its business partners through supply chains.