Year of Graduation
Modeling of the Oil Price Formation Process in the Global Oil Market
School of Statistics, Data Analysis and Demography
The following paper is devoted to the problem of oil price formation process modeling. The main aim of the study is to conduct an in-depth analysis of the oil price formation process and suggest a new model of the price formation in the global oil market.The author first provides the historical reference of the world oil market development: special characteristics and major tendencies of the oil market are determined. It is proved that in spite of the growing impact of financial markets on the oil price formation, structural aspect still matters, because all historical price shocks can be explained mainly by supply and demand factors, therefore they should be necessary included in the model of oil price formation. Also, it is stated that oil market significantly depends on the macroeconomic and political situation in the most developed countries.The second section is devoted to the results of econometric analysis. The analysis conducted by the Author includes trend, seasonal, adaptive and autoregressive models. It is proved that standard methods of time-series analysis, such as (autoregressive moving average) ARIMA model or adaptive Brown’s model, may be used for oil price modeling. In the third section the Author offers a new mixed model of the oil price formation, that takes into account both structural and financial factors of the oil market. The proposed model is based on the academic papers by Dees (2007) и Kaufmann (2010), who focus mainly on the structural factors of the oil market and by Ellen (2010) и Andreopoulos-Loukazevits (2011), whose studies are suitable for “oil as a financial asset” side. The structural part of the model, proposed by the Author, is based on the finding of market equilibrium using model by Hubbert (1956) and microeconomic principle of corner solutions for supply estimation and regression analysis for oil demand simulation. As for financial side of the model, it is simulated on the base of heterogeneous autoregressive model (HAM) according to agent’s theory and principles of behavior economics.