Economic Data Analysis
Our first group project designed to highlight use/knowledge of Python, Pandas, loading data from CSV and APIs, creating visualizations in MatplotLib.
As one of our group members works in the oil field services industry, we took a look at economic data that could explain changes in employment and wages in the oil field.
We used data from the Federal Reserve Board of St. Louis, Consumer Financial Protection Bureau, Bureau of Labor Statistic, and the US Energy Information Administration
We found that changes in the overall employment level in the US preceeded changes in average hourly earnings, though the leadtime varied greatly, making precise estimates of changes in average hourly earnings difficult. Changes in long-term interest rates seemed to be more coincident with changes in hourly earnings.
We then shifted our focus to looking at employment in the oil field service area. We used the Texas unemployment rate as our proxy, and compared that to the NYC unemployment rate, as that is the economic region with which we are most familiar.
Unemployment peaked in both regions in 2010, after the Great Recession. While unemployment has continued to decrease in NYC, there was an increase in unemployment in Texas in 2016, with a slight decrease in 2017.
Overlaying a chart of oil prices, we found that the collapse in oil prices in 2015 and 2016 was correlated with the increase in unemployment in 2016. As oil prices have recovered somewhat, unemployment in Texas has resumed its decline. Hopefully the economy in Texas has diversified away from oil in recent years.
The Powerpoint slideshow is in Final_AverageCoders.pptx