A UC Berkeley Project done for For Data 100 Fall 2021
Seeing Covid-19 data and its correlation with Unemployment rates throughout counties in the U.S.
The model used was a linear regression model. This was an appropriate model to use because we want to use continuous quantitative variables: case numbers and masking rates, to predict a continuous quantitative variable: the amount of unemployed people in a county. The inputs to the model are case numbers and masking rates in each county, while the output is the amount of unemployed people in a given county.
Furthermore, A/B testing was preformed to show unemployment rates went up due to covid-19 along with some visualizations on the distribution of unemployment rates in 2018 (pre covid) and 2020 (covid).
The result can be seen in the Jupyter Notebook!!