2021 Fiscal Year Final Research Report
Achieving real-time nowcasting of corporate performance and stable stock markets using POS data
Project/Area Number |
17K01277
|
Research Category |
Grant-in-Aid for Scientific Research (C)
|
Allocation Type | Multi-year Fund |
Section | 一般 |
Research Field |
Social systems engineering/Safety system
|
Research Institution | Kanazawa Gakuin University |
Principal Investigator |
|
Project Period (FY) |
2017-04-01 – 2022-03-31
|
Keywords | POSデータ / ナウキャスト |
Outline of Final Research Achievements |
In order to develop a statistical technology for calculating real-time business performance from POS data, we linked the performance of listed companies by Reuters to a database of POS data from about 5% of supermarkets in the United States. Next, from the correlation between POS sales and actual sales, we confirmed that the growth rate of actual sales of companies whose main products are daily necessities is strongly correlated with the growth rate of POS sales, and that the correlation between the growth rate of POS sales and the growth rate of actual sales becomes stronger as the ratio of sales of supermarkets to total sales of companies (called the coverage ratio) is larger. From these results, we conclude that for companies with a high coverage ratio, it is possible to nowcast actual sales in the next quarterly settlement with prediction accuracy from the comparison of POS sales from the previous period and the same period of the previous year.
|
Free Research Field |
経済物理学
|
Academic Significance and Societal Importance of the Research Achievements |
株価は、企業の業績を背景に、100 分の1 秒の間隔で変動する。一方、企業の業績は、四半期に一度しか公開されない。そのため投資家は、直近の公開情報から企業の業績を予測するしかなく、時として、他者の行動に振り回される。本研究成果により、企業の業績が常にリアルタイムで算出されれば現在の業績を予測する必要がなくなり、予測期間は大幅に短縮される。その結果、株価のフラッシュクラッシュは抑えられ、四半期に一度の業績公開による株価の乱高下は消滅し、安定した株式市場が実現される。
|