Research on the impact of information interaction of investor interaction platform on the foam of listed companies'stock prices:Evidence based on text deep learning
This paper builds a parallel CNN-LSTM deep learning model,excavates the interac-tive text between investors and the management of listed companies on the interaction platform of Shanghai and Shenzhen Stock Exchanges,quantifies the information interaction process be-tween investors and the management from the text content and semantic characteristics,and analyzes its impact on the company's stock price foam by identifying the content of investors'inquiries and the quality of company responses.The study found that investors'questioning and attention to different contents had significantly different effects on the stock price foam.Among them,questioning and attention to the company's stock and financial information could inhibit the frequency and intensity of the stock price foam,while questioning and attention to the company's research,development,production and sales information increased the frequency and intensity of the stock price foam;The higher the clarity of the management's response to investors'questions,the more significantly it can reduce the frequency of stock price foam,but it has no significant impact on the strength of stock price foam.Further,the worse the external information environment is and the lower the shareholding ratio of institutional investors is,the more obvious the impact of information interaction on the foam of the company's share price is.The transparency of company information plays a part of the intermediary effect in this impact.The research findings contribute to a micro perspective understanding of the impact mechanism of information exchange behavior on the operation of the stock market in China's investor inter-action platforms.
NETL
NSTL
万方数据
