Download PDFOpen PDF in browser

Correlating Social Media Sentiment with Stock Market Volatility Exploring Relationships Between Sentiment and Market Fluctuations

EasyChair Preprint 14881

26 pagesDate: September 14, 2024

Abstract

The increasing influence of social media on financial markets has created new opportunities for understanding stock market volatility through sentiment analysis. This paper explores the correlation between social media sentiment and fluctuations in stock prices, investigating whether sentiment extracted from platforms such as Twitter, Reddit, and StockTwits can serve as a predictor for market volatility. By employing natural language processing (NLP) techniques to analyze public sentiment and correlating this data with market volatility indices, we aim to determine if significant relationships exist between public opinion and financial market behavior.

 

Through case studies of events such as the rise of meme stocks (e.g., GameStop, AMC) and the volatility of cryptocurrencies (e.g., Bitcoin), we illustrate how social media sentiment has influenced market dynamics. Results from correlation analysis show that shifts in public sentiment can precede or coincide with stock price movements, with certain sentiment trends (e.g., panic or optimism) having stronger predictive power in volatile market conditions. The study highlights the potential for investors and traders to use sentiment analysis as a tool for anticipating market swings while acknowledging the limitations of relying solely on social media data.

Keyphrases: Market Behavior, Sentiment Analysis, correlation analysis, predictive modeling, social media sentiment, stock market volatility

BibTeX entry
BibTeX does not have the right entry for preprints. This is a hack for producing the correct reference:
@booklet{EasyChair:14881,
  author    = {Docas Akinyele and David Ray},
  title     = {Correlating Social Media Sentiment with Stock Market Volatility Exploring Relationships Between Sentiment and Market Fluctuations},
  howpublished = {EasyChair Preprint 14881},
  year      = {EasyChair, 2024}}
Download PDFOpen PDF in browser