Measurement of Investment activity in China based on Natural language processing technology
Xiaobin Tang,Tong Shen,Manru Dong
DOI: https://doi.org/10.48550/arXiv.2211.16829
2023-04-05
Abstract:The purpose of this study is to propose a new index to measure and reflect China's investment activity in time, and to analyze the changes of China's investment activity in the past five years. This study first uses the NEZHA model for semantic representation, and expand the indicator system based on semantic similarity. Then we calculate China's investment activity index by using the network search data. This study shows that China's investment activity began to decline in 2019, rebounded for a period of time after the outbreak of COVID-19 in 2020, and then continued to maintain a downward trend. Private investment activity has declined significantly, while government investment activity has increased. Among the provinces in Chinese Mainland, the investment activity of economically developed provinces has decreased significantly, while the investment activity of some economically less developed provinces in the north and south is higher. After the outbreak of COVID-19, the investment period became shorter. Our research will provide timely investment information for the government, decision makers and managers, as well as provide other researchers who also pay attention to investment with a perspective other than investment in fixed asset.
Applications