Simulating Rubber Prices under Geometric Fractional Brownian Motion with Different Hurst Estimators

Abstract

Natural rubber was a vital pillar of Malaysia's export-oriented economy throughout much of the twentieth century, according to the Economic History of Malaya (EHM) website, the worldwide demand for natural rubber is expected to expand at a CAGR of 4.8% in the future (2019–2023). Therefore, it could be concluded that rubber is one of the commodities that should be monitored. Thus, the forecasting of the rubber price should be analysed for the future use. This research focuses on the simulation of rubber prices using geometric Brownian motion and geometric fractional Brownian motion. Moreover, an analysis in the value of hurst exponents' estimator were also done using 3 different methods, Rescaled Range Analysis, Aggravate Variance Method, and Residuals of Regression Method that bases on the slope deviation technique. Finally, an error analysis was done by the calculation of Mean Absolute Percentage Error (MAPE) value to analyse the accuracy of the simulation models.