Performance Comparison between Geometric Brownian Motion and Heston Model for Assessing Silver Price Risk under Thai Baht to US Dollar Exchange Rate Volatility

Abstract

This study aims to 1) compare the performance of the Geometric Brownian Motion model and the Heston Model in assessing silver price risk, 2) analyze the relationship between global silver prices and the Thai baht–US dollar exchange rate (THB/USD), and 3) evaluate tail risk in the context of dual risk exposure. The significance of this research is reflected in the Thai jewelry industry, which relies heavily on silver imports with an average annual value exceeding 50 billion baht. The study employs daily secondary data covering    a five-year period (2021–2025) obtained from the Yahoo Finance database, using the closing prices of silver futures contracts (SI=F) and the Thai baht–US dollar exchange rate (THB=X). Monte Carlo simulation with 5,000 iterations was then conducted to analyze the price distribution and evaluate tail risk in Thai baht terms. The results indicate that the Heston Model, which incorporates stochastic volatility, performs better than the Geometric Brownian Motion model in capturing fat-tail behavior, producing a kurtosis value of 6.31, compared with 3.87 for the Geometric Brownian Motion model. In addition, rolling correlation analysis reveals a short-term positive relationship at r = 0.287, suggesting that the natural hedging mechanism does not operate effectively under current market conditions. Furthermore, the Heston Model provides a more comprehensive estimation of losses under worst-case scenarios. Based on these findings, the study recommends that businesses consider adopting a decoupled hedging strategy to maintain the stability of import costs under conditions of heightened market volatility.