Do people express and pay more attention to positive or negative information in stock forums? A big data analysis

Abstract

Investment decision-making is an important and challenging task faced by people in modern society, and emotions play an important role in the process of investment decision-making. Existing research mostly focuses on predicting the stock market trends based on the sentiment of posts in stock forums, but no research has yet taken an overall perspective to study the emotional tendencies in how people express and focus on information in stock forums. Positivity bias and negativity bias are two opposing theories, under different backgrounds and conditions, people may exhibit either a positivity or negativity bias, but in stock forums, it remains unclear which of these emotional tendencies predominates. This study aims to fill this gap by using big data analysis, dictionary-based natural language processing, and engagement indicators to analyze nearly 1.3 million user posts from East Money Stock Bar, the largest stock forum in China. We calculated the count and intensity of sentiment word usage to explore the phenomenon of emotional bias in stock market discussions. The results indicate that whether expressing or focusing on information, a investors’ emotional performance in stock forums tends to be positivity bias. This finding provides an effective supplement to the positivity bias theory, helping us to gain a deeper understanding of people’s psychological characteristics in the investment decision-making process, and has certain theoretical value; at the same time, this study provides support for investors to analyze market information more scientifically, and has practical significance for using emotions to predict the stock market.

Keywords

• Positivity bias
• negativity bias
• stock forums
• sentiment analysis
• big data psychology