Examining Cryptocurrency Price Movements and Twitter Sentiment Using VADER and Econometric Models

Abstract

The volatility of cryptocurrency markets has attracted significant attention, with social sentiment emerging as a potential factor influencing price movements. This study investigates the relationship between cryptocurrency prices and Twitter sentiment, focusing on Bitcoin and Ethereum. The purpose of this research is to examine the bidirectional dynamics between market sentiment, derived from Twitter data, and cryptocurrency price fluctuations. A sentiment analysis using the VADER tool is coupled with econometric models, specifically the Vector Autoregression (VAR) and Granger Causality tests, to analyze price and tweet data from May 2017 to May 2019. The results reveal a bidirectional relationship, where price changes influence sentiment, but sentiment plays a limited role in driving price movements. Specifically, while Twitter sentiment reacts to price shifts, it does not significantly predict future price changes. The Granger Causality tests confirm that price trends are the primary drivers of market behavior, while sentiment mainly reflects public reactions to these trends. The findings suggest that investors should prioritize price data and market fundamentals over sentiment analysis for decision-making, as sentiment alone lacks predictive power in this context. This research contributes to the understanding of cryptocurrency market dynamics, offering insights for investors and policymakers in navigating the complexities of digital asset markets.