Crude oil prices typically maintain certain statistical relationships with the products refined from individual crude oils. These relationships can be described using many standard multivariate techniques, among them correlation and Principal Components Analysis. In addition, times series techniques can be used to describe the behavior of the individual crude and product prices. A project in this area would assess price relationships for US crudes (e.g. WTI, Kern River) refined in the US against their refined products and then perform similar local comparison for Nigerian crudes (e.g. Bonny Light, Qua Ibo) and North Sea crudes (e.f. Brent Blend, Ekofisk)

• Identify appropriate crude oil and product prices in at least three separate locations.
• Collect price data for the crudes and products and check the data for missing observations and bad data.
• Perform Exploratory Data Analysis on the data including time series plots, Q-Q analysis.
• Compare key statistical metrics across the different markets and assess similarities and differences, including measures like variance, skew, kurtosis, covariance, ARIMA parameters.
• Discuss possible economic explanations for differences.

These instructions will get you a copy of the project up and running on your local machine for development and testing purposes. See Deployment / Installation for notes on how to deploy the project on a live system.

• Operating System – Microsoft Windows 10
• Python Version – Python 3.7
• Python Modules
  • pandas
  • matplotlib
  • scipy
  • stats
  • statsmodel.api
  • seaborn
  • sklearn

Visit https://www.microsoft.com/software-download/windows10 for instructions on installing Windows 10 operating system.

It is advisable that the Python modules are installed using Anaconda package. Please visit https://www.anaconda.com/distribution/ for further instructions on installation.

The Python code for the Project can be found in the Github address below:

https://github.com/tarunk/CAPSTON

Load the code into the Python environment (Anaconda IDE) and run as necessary.