Analysis of Supply and Demand Dynamics to Predict Oil Market Trends: A Case Study of 2015 Price Data

Abstract

Abstract Understanding the oil price changes through the demand and supply dynamics is a challenging problem. Traditional models based on stochastic process and aggregated supply and demand have been successful in explaining the price trends in short periods of time where there is no big changes in either supply or demand sides of oil market. However, these models usually failed to capture the big shocks of oil price in circumstances where there are unexpected influential events, such as reports on the economic growth of the non-OECD members. In this paper, we propose a revised model of oil price based on the supply and demand dynamics. Particularly, we inspect the effects of unconventional parameters on the oil price, many of which have been ignored in the traditional models. Our model investigates the effects of these expectational parameters on the major factors that shape the supply and demand. Particularly, the oil price is still modeled to follow the supply to demand ratio, however, the supply and demand are replaced with their expected trends in the model. In other words, based on the influential events that can happen in either demand or supply sides, related variables are initiated which subsequently affect the expected trend of supply or demand, and the oil price eventually. Our model is a system dynamic model that mathematically relates the determinants of the oil price based on the underlying causal relations. The whole model numerically solves differential equations relating the marginal parameters and the main factors including the oil price, demand and supply. We separate the oil supply into US, OPEC and other non-OPEC countries supplies whose parameters differ due to conventional and unconventional resources. Moreover, we separate the oil demand into OECD, Brazil, Russia, India, China (BRIC) and non-OECD countries’ demands due to the difference in the dependence on the economic growth and oil consumption in those group of countries. We build our core model based on regression analysis on the historic data of WTI oil price, OECD and BRIC economic growth and demand, US, OPEC and other countries supply data. Then we train other parameters of the model using the historic reactions of the market to unexpected events that have occurred in the past. Finally, we simulate the model to analyze a case study of the events occurred in the 2015 including the deceleration of the growth in Chinese economy, speculations on American shale productions, speculations on the OPEC’s decisions, and conflicts in the Middle-East region. Our results show that our model can be used to reproduce the price trends compliant with the past data, and provide predicted trends of supply, demand and their influences on the price.