Simulating rubber prices under geometric fractional Brownian motion with different Hurst estimators

This research focuses on the simulation of rubber prices using geometric Brownian motion (GBM) and geometric fractional Brownian motion (GFBM). An analysis of the Hurst values estimated using three different methods, namely, rescaled range analysis, aggravate variance, and residuals of regression. These methods are based on the slope deviation technique. An error estimation via Mean Absolute Percentage Error (MAPE) is computed to analyse the accuracy of the simulation models which shows that the GBM is highly accurate over a 120-day period, and as the number of days increases, the accuracy decreases. Moreover, the rescaled range analysis method can estimate the Hurst values for all five ranges of days in comparison to the residuals of regression and the aggregate variance methods. The rescaled range analysis method also able to produce a highly accurate simulation up to 150-day.

Actions (login required)

View Item