Forecasting RMB exchange rate volatility: do time-varying higher moments and time-varying risk aversion help?

ABSTRACT

In this paper, we investigate the predictive value of time-varying higher moments and time-varying risk aversion (RA) for the RMB exchange rate volatility. To do so, we develop a threshold GARCH-MIDAS model with skewness and kurtosis (henceforth TGARCH-MIDAS-SK model). Our empirical results indicate the presence of significant reverse leverage effect and time-varying skewness and kurtosis in the RMB exchange rate returns. The RA has a significant negative impact on the RMB exchange rate volatility. Moreover, we observe that incorporating leverage effect, time-varying higher moments and RA improves the in-sample fitting and out-of-sample forecasting performance of the model.

KEYWORDS:

• RMB exchange rate volatility
• time-varying higher moments
• time-varying risk aversion
• TGARCH-MIDAS-SK model
• leverage effect

JEL CLASSIFICATION:

• C5
• C32
• G17

Acknowledgements

We would like to thank the Editor, Mark Taylor, and two anonymous referees for their valuable comments and suggestions that greatly improved the paper.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

¹ As suggested by a referee, the IGARCH model which is able to accommodate the non-stationarity in the conditional variance process should be considered as a benchmark.

Additional information

Funding

This work was supported by the National Natural Science Foundation of China under [Grant No. 71971001]; Natural Science Foundation of Anhui Province Excellent Youth Scientists Fund Project under [Grant No. 2208085Y21]; Academic Funding Project for Top Academic Talents in Universities in Anhui Province under [Grant No. gxbjZD2022019]; Anhui University Collaborative Innovation Project under [Grant No. GXXT-2021-078]; University Research Project for Graduates of Anhui Province under [Grant No. YJS20210420]; Innovative Research Project for Graduates of Anhui University of Finance and Economics under [Grant No. ACYC2020185].