A Predictive Model for Oil Market under Uncertainty: Data-Driven System Dynamics Approach

TL;DR

This paper introduces a data-driven system dynamics model that predicts oil prices under uncertainty by incorporating expectational variables, improving upon traditional models that fail during rapid market changes.

Contribution

It presents a novel system dynamics approach that integrates expectational variables into oil price prediction, validated through historical scenario simulations.

Findings

Model accurately replicates past oil price trends.

Incorporates expectational demand and supply variables.

Simulations align with historical market scenarios.

Abstract

In recent years, there have been a lot of sharp changes in the oil price. These rapid changes cause the traditional models to fail in predicting the price behavior. The main reason for the failure of the traditional models is that they consider the actual value of parameters instead of their expectational ones. In this paper, we propose a system dynamics model that incorporates expectational variables in determining the oil price. In our model, the oil price is determined by the expected demand and supply vs. their actual values. Our core model is based upon regression analysis on several historic time series and adjusted by adding many casual loops in the oil market. The proposed model in simulated in different scenarios that have happened in the past and our results comply with the trends of the oil price in each of the scenarios.