Multivariate Dataset For Base Oil Forecasting

This Multivariate dataset is composed of 386 variables for base oil forecasting and it was developed using financial market data from Yahoo Finance website. The variables are composed of global stock indexes, currency, metals, oil derivates, and companies from the following sectors: refineries, energy, oil, gas, lubricants, oil derivates (gasoline, diesel, etc.), pipelines, oil tankers, exploration and production (E&P), gas station, EPCM (wich operates in the oil & gas sector).

Many variables are traded in other currencies besides US Dollar. Therefore, it was necessary to capture the relationship of all these currencies to the US Dollar. Each currency ticker reflects this exchange relationship. This was necessary to standardize all prices in US Dollar.

The metals tickets are related to Oil as they can be used as a financial hedge against Oil volatility. So, these tickers are variables related to Oil.

US Treasury tickers were used as they are also used as financial hedges by portfolio managers to counterbalance the Oil and markets volatility.

The period considered is between 2010-01-01 and 2025-10-31.

The spreadsheet contains two tabs. The first is the Multivariate dataset and the second is a set of tickers. There were no captured because the prices captured had less than 80% of the period covered or because the ticker was not found.

The values captured were the Adjusted Close Prices that represent a stock's closing price modified to account for all applicable corporate actions, including dividend distributions, stock splits, and rights offeringsadjusted for stock splits.