Author(s)

Yuan Lu, Zhurong Geng

Corresponding Author

Yuan Lu

ABSTRACT

This paper discusses the relationship between investor sentiment and the realized volatility of gold futures, which is investigated using high-frequency data. For investor sentiment factors, we select four indicators including the volatility index (VIX), the volume, inventory and turnover rate of gold future. To improve our forecasting accuracy, stock market factor is introduced and we select S&P500 index to represent it. Based on the heterogeneous autoregressive (HAR) theory, six new heterogeneous autoregressive (HAR) models are established by combining investor sentiment and S&P500 index. The empirical results show that the accuracy of the new model is better than that of the original HAR model. We found that S&P500 index contain a lot of gold prediction information. In addition, the investor sentiment has a positive impact on the volatility of gold futures. Our work is the first to combine investor sentiment with the S&P500 index to identify more market information. This paper provides a better forecasting method for the Volatility Prediction of gold futures.

KEYWORDS

Investor sentiment, HAR-type models, Gold futures market, Volatility forecasting