Interaction between Oil Price and Investor Sentiment: Nonlinear Causality, Time- Varying Influence, and Asymmetric Effect

ABSTRACT

This paper investigates the interaction between crude oil prices and individual investor sentiment with the Hiemstra and Jones (HJ) test, the Diks and Panchenko (DP) test, the time-varying parameter structural vector autoregression (TVP-SVAR) model, and the nonlinear autoregressive distributed lags (NARDL) model. Results reveal a bidirectional nonlinear Granger causality, rather than a linear Granger causality, between crude oil prices and individual investor sentiment. Meanwhile, the interactions between the two variables are time-varying, and oil prices negatively affect individual investor sentiment in general. However, the effect of individual sentiment on oil prices is more complicated. It has more significant impacts on oil prices after 2000, and shows a positive influence before the global financial crisis, a minor influence during the crisis, and even a negative influence after the crisis. In addition, the oil price has significant long-run and short-run asymmetric effects on individual investor sentiment, whereas individual investor sentiment has no asymmetric effect on oil prices.

KEY WORDS:

• asymmetric effect
• investor sentiment
• oil price
• time-varying influence

Additional information

Funding

This work was supported by the National Natural Science Foundation of China (Nos. 71701081, 71673277) and the Fundamental Research Funds for the Central Universities (No JUSRP11887).