Does social network sentiment influence the relationship between the S&P 500 and gold returns?

• Published in: International review of financial analysis Vol. 57; pp. 57 - 64
• Main Authors: Piñeiro-Chousa, Juan, López-Cabarcos, M. Ángeles, Pérez-Pico, Ada María, Ribeiro-Navarrete, Belén
• Format: Journal Article
• Language: English
• Published: Elsevier Inc 01.05.2018
• Subjects: ARIMA; GARCH; Gold; S&P 500; Social media sentiment; Gold; GARCH; S&P 500; ARIMA; Social media sentiment; G10
• ISSN: 1057-5219; 1873-8079
• DOI: 10.1016/j.irfa.2018.02.005

This study explored the relationship between investor sentiment (extracted from the StockTwits social network), the S&P 500 Index and gold returns. We investigated bilateral causality between gold prices and S&P 500 prices, the power of investor sentiment and gold returns to predict S&P 500 returns, and the influence of gold returns on S&P 500 volatility. We also considered whether the influence of sentiment varies according to the user's degree of experience. We considered the sentiment of messages that mentioned the S&P 500 Index and that users posted between 2012 and 2016. Granger causality analysis, ARIMA models and GARCH models were used for predicting S&P 500 Index returns and S&P 500 volatility. We observed a causal relationship between gold price and the S&P 500 Index. Our results also suggest that sentiment and gold returns predict S&P 500 Index returns. Finally, we observed that gold returns influence S&P 500 volatility and that the sentiment of experienced users affects S&P 500 returns. •There is a bilateral causality between gold price and S&P 500.•Investor sentiment and gold returns can predict S&P 500 returns.•Gold returns influence the volatility of S&P 500 returns.•Sentiment from experienced investors influences S&P 500 returns.