Dynamics of Conditional Causality Between Bitcoin Price and Social Sentiment

Bitcoin price action and volatility is driven by various factors, including social sentiment. Several studies have explored the peer-to-peer causal relationships of statistical behaviors between social sentiment and Bitcoin price actions. However, the conditional dynamic interdependencies of statistical features of social sentiment and Bitcoin has not been examined. This paper presents an analysis framework to examine such conditional dynamics. We explore the patterns of information exchange among these statistical behaviors across various time scales, and extract and analyze the dynamic rules associated with their conditional influence on each other. Our method reveals patterns of complex conditional communications between social sentiment and Bitcoin price under different scenarios. Specifically, we find that bullish posts have a more enduring and substantial impact on investors' decisions compared to negative posts. Moreover, sentiment trends exhibit stronger predictive power for the near future than for the distant future. Additionally, our findings reveal a significant correlation between heightened levels of irregularity in the trend of Bitcoin or social sentiments and a subsequent decrease in the future value of Bitcoin. This research contributes to a better understanding of the interdependencies among financial time-series characteristics, providing valuable implications for investment decision-making and outcome prediction in the realm of digital currencies.

کلیدواژه ها:

Time-varying causality ، Dynamic rules ، Conditional inference ، Bitcoin ، Cryptocurrency ، Sentiment analysis