Studies On The Dynamic Relationships Between Investor Sentiment And Soybean Futures Returns In The Context Of Sino-US Trade War

With the rapid growth of China's economy and the improvement of people's living standard,spot investments have been unable to meet the needs of investors,and investors began to shift their attentions from the spot market to futures and other financial markets.At present,investor sentiment has become one of the indicators that investors pay more attentions to.From the early structured index to the text with implicit emotional semantics,it has attracted the attentions of many financial enterprises and regulators and has gradually become one of the hot issues studied by scholars.With the development of futures market in recent years,many researches show that there remain a correlation between investor sentiment and futures returns,but the conclusions are not uniform.On March 22,2018,the trade war between China and the United States started.The United States exerted trade pressure on China,and China imposed tariffs on imported soybeans and other commodities from the United States.With the advent of the era of big data,it is particularly important to study the dynamic relationship between investor sentiment and futures returns through text mining technology.On the basis of previous research results,this paper explores the dynamic correlation between investor sentiment and soybean futures returns in the context of Sino-US trade war,so as to provide theoretical basis and reference for investors to trade.Mainly from the following aspects:1.Investors' cognition of the trade war between China and the United States usually comes from TV or mobile terminal,and mobile terminal devices such as computers and mobile phones gradually occupy people's lives and become the carrier of texts that directly affect investors' emotions.Through the collection of WIND news on July 2,2017(solstice,October 15,2019),this paper uses text word segmentation,filtering stop words,and extracting text keywords to denoise the text of the trade war news to remove emotionless words and improve the accuracy of emotion quantification.The SnowNLP module quantifies the news text into the mood index,and differentiates the news text from the positive mood news text and the negative mood news text,and GARCH(1,1)model is used to depict themood fluctuation.2.Through the analysis of the high frequency words,semantic network analysis,LDA model and descriptive statistical methods to explore the main information,the logical relationship between the main information of news,news text features and the theme of the news topic and the relationship between the investor sentiment characteristics,the conclusion shows that compared with the positive news,when negative news dominated,investors are more significantly mood moves.3.By descriptive statistical method and GARCH(1,1)model to explore general information of the soybean futures returns and the law of the soybean futures returns volatility,the conclusion remains that the soybean futures returns presented a non-normal distribution of the right in line with the "aggregation" characteristics of financial data,and futures returns volatility by historical futures returns the positive influence.4.DCC-GARCH model was established by using the returns rate and sentiment index to explore the dynamic relationship between investor sentiment and soybean futures returns in the context of the Sino-US trade war.The results shows that the dynamic correlation between investor sentiment and soybean futures returns in the context of the Sino-US trade war changed from stable fluctuation to negative correlation,then back to positive correlation and gradually stabilized.