Hey, What’s Up? Gas Prices: Analyzing the Influences of U.S. Gas Price Trends

Jennifer, with her teammates, explored the factors that affected gas prices. Society had a lot of hypotheses for why there were such large fluctuations in commercial gas prices during the recent Russia-Ukraine conflict and the COVID-19 pandemic. The team wanted to find the root reasons behind the rise and fall of gas prices.

To begin with, the team did research on the supply chain for commercial gas such as importing, refining, and transportation. Then, they created their own dataset by merging multiple sets from six different sources and cleaned it. The resulting dataset included variables such as average gas cost, imports, and renewable energy use.

After the data cleaning and exploratory analysis, the team used a variety of data analytics tools learned from the Data Science Academy hosted by Wharton. They applied text mining, LASSO regression, and random forests to determine the most influential factors that affect gas prices. They found that the amount of imported barrels, total consumption of fossil fuels, CO2 emissions from various sectors, percentage of U.S. emissions in total global emissions, and the percentage of wind in total renewable energy consumption were the leading factors in gas price changes.

The team was invited out of 15 groups to present at Women in Data Science @Penn on February 3rd.