A novel term structure stochastic model with adaptive correlation for trend analysis

The prediction of underlying price continues to draw extensive attention in academic research. Based on a review of the advantages and disadvantages of different volatility models, we find that the Heston stochastic volatility model with an adaptive correlation coefficient is most suitable for analysing the Hong Kong options market. We subsequently propose a model-free implied volatility term structure formulated using options with different strikes and different maturities. The implied volatility is calculated by integrating the option price and strike price from the current time to the expiry date. Discrete points of term structure data are used to fit a term structure curve. Finally, we use the model-free implied volatility term structure as the long-run mean level of the Heston model to fully exploit the information content contained in the implied volatility term structure. We simulate the distribution of the underlying asset price based on the Heston model and constant elasticity of variance (CEV) model. The adaptive correlation Heston model provides superior results in terms of one-day-ahead prediction performance and the 79-day distribution of the underlying asset price compared with the CEV model.