VAR & VECM Brent Crude Oil Forecast January 2021-June 2021

Crude oil is the lifeblood of the industrialized nations. Oil has become the world's most important source of energy since the mid-1950s. Its products underpin modern society, mainly supplying energy to power industry, heat homes, and provide fuel for vehicles and airplanes to carry goods and people all over the world[1]

Crude oil also creates petroleum products. [2] Petroleum by-products make tar, asphalt, paraffin wax, and lubricating oils.[3] It is also used in chemicals, such as fertilizer, perfume, insecticides, soap, and vitamin capsules.[4]

World crude oil supplies are from many benchmarks which mainly dominated by Brent, OPEC Reference Basket, Dubai Crude, Oman Crude, Shanghai Crude, Urals oil and West Texas Intermediate (WTI). But among the world benchmarks, the crude oil Brent is the leading global price benchmark for Atlantic basin crude oils. It is used to set the price of two-thirds of the world's internationally traded crude oil supplies. Supporting the importance of Brent in the global pricing of crude oil, commodity indices have increased their allocation to Brent. Historical price trends show that Brent is closely correlated to other international grades such as LLS, Dubai, and Mars.[5]

Thereafter, knowing the importance of crude oil, especially Brent, I am trying to understand the behavior influence factors and to provide the price forecast for the next 6 month periods (January to June 2021) with two model prediction provided which are the Vector Auto-Regressive Model (VAR) and the Vector Error Correction Model (VECM).

On the result, the VAR and the VECM show different estimations*. The VAR model indicates an upward trend estimating the Brent will average 61.99 $/Bbl on June 2021 with an escalating increase in February 2021. While the VECM indicates that Brent will experiences an overall downward trend with a one-time increase in April 2021.

For comparison here are some other reliable forecast for Brent Future Price in 2021

STATISTICAL OVERVIEW

On the methodology, knowing that the Brent prices have fluctuated and have many uncertainties I chose to forecast the Brent by using the Vector Auto-Regressive Model (VAR) and the Vector Error Correction Model (VECM).

For this research purpose, I used many factor groups of variables to try to understand what variables influenced the movement of the Brent prices with the data are covered on monthly average periodicity from January 2018 to December 2020. The over mentioned factor groups are Economics; Fundamental Market; Exploration Technology; Financial Market; Speculation; and Geopolitical. Based on my preliminary variable tests (Unit-root, Lags, and Cointegration Tests) I decided to choose the logarithmic form of the Brent Prices($/barrel); OECD countries’ oil stocks(day of supply), World Oil Consumption (Million Barrels per day), and the Dow Jones Industrial Average indexes for the variables. **Disclaimer**, I think the Brent price is influenced by the world geopolitical and COVID-19 condition, but I cannot find sufficient statistical proof on that hypothesis. Thereafter, I put the Dow Jones Industrial Average Index that was influenced by those factors and has good statistical proof on correlating to the movement of the Brent.

To proof the explanatory variable parameters are statistically reliable for the forecast, I have implemented some pre-estimation and post-estimation diagnostic tests, which are:

Pre-estimation*

1.      ADF Unit Root Test for Stationary;

2.      Optimal Lag Order Determination;

3.      And, Johansen Co-Integration Test.

Post-estimation*

1.      Granger Wald Test for Causality;

2.      La-Grange Multiplier Test for Auto Correlation;

3.      Wald Test for Coefficient Short-run Casualty;

4.      Jarque-Bera Test for the checking error normal distribution;

5.      And the Eigen Stability test.

ACKNOWLEDGEMENT

This research was done independently based on my personal interpretation and calculation, I apologize if there is some misinterpretation or wrongly used methodology. I want to dedicate this result of the forecast for The MIT Center of Transportation and Logistics  and many thanks to Prof Chris Caplice, Ms. Eva Ponce and the Teaching Assistants Jalaluddin, Sindhu Srinath and the others that are not mentioned here that has taught me on forecasting basics and also much other knowledge on Supply Chain and Logistics on the past SC0x and SC1x courses, I am looking for another great journey on the SC2x Micromasters course coming on January 6. Also I want to thank Ms.Bosede Ngozi Adeleye for her Crunch Econometrics and M Dzaki Fahd Haekal for helping me understanding further about VAR VECM and STATA.

[1] UKOG, “Why Oil Is Important”

[2] U.S. Energy Information Administration. "What Is the Difference Between Crude Oil, Petroleum Products, and Petroleum?

[3] U.S. Energy Information Administration. "Glossary”

[4] The Balance, “Crude Oil, Its Types, Uses, and Impact”

[5] Wikipedia, “Brent Crude”