| dc.description.abstract | Global crude oil price volatility represents a major source of risk for energy
companies, particularly oil refining firms that rely heavily on crude oil imports and
face limitations in passing cost increases to end consumers due to price control
policies. PT XYZ, as a national oil refining company, is exposed to significant crude
oil price risk, where fluctuations in global oil prices directly affect import costs,
refining margins, and overall financial performance. Although commodity hedging
strategies have been implemented, their realization remains limited and largely
reactive, relying on short-term technical signals and therefore lacking effectiveness
in reducing import cost volatility.
This study aims to examine the application of the Ornstein–Uhlenbeck (OU)
stochastic model as a quantitative framework to support crude oil hedging decisions
at PT XYZ. The research focuses on evaluating the ability of the OU model to
project crude oil prices, determine risk-based hedge ratios, and assess the
effectiveness of an OU-based hedging strategy compared to the company’s existing
hedging approach.
The study employs monthly Dated Brent price data with observation horizons
of three, five, and ten years. OU model parameters are estimated using the
Maximum Likelihood Estimation method, followed by Monte Carlo simulations to
generate six-month forward price distributions in the form of P5, P50, and P95
quantiles. Based on validation results using Root Mean Square Error, Mean
Absolute Error, and the coefficient of determination, the five-year horizon OU
model is selected as the most representative due to its balance between statistical
accuracy and sensitivity to current market conditions.The results indicate that crude
oil prices exhibit strong mean-reverting behavior, confirming the suitability of the
Ornstein–Uhlenbeck model for short- to medium-term price projection. OU-based
price projections provide median price estimates converging toward the long-term
equilibrium level, along with measurable uncertainty ranges. Integrating these
projections with the company’s risk tolerance enables the determination of
proportional and risk-based hedge ratios, ensuring that hedging decisions account
not only for price direction but also for maximum potential loss control.
The main contribution of this study demonstrates that an OU-based hedging
strategy reduces the volatility impact of crude oil price fluctuations more effectively
than the existing strategy and produces more consistent and replicable decisions.
The Ornstein–Uhlenbeck model is not intended to replace the existing hedging
strategy operationally, but rather to complement it by strengthening the decision-
making framework with a quantitative, risk-based foundation. Practically, this
approach supports the development of a more systematic, proactive, and
governance-aligned risk-based hedging framework, thereby enhancing commodity
risk management at PT XYZ. | |
