darya-akimova
commented
6 years ago Looked through these notebooks and it seems like some interesting work was done. A problem is that the dependent variable (medicare spending) and independent variable (lobby spending) look like they were switched when the model was built (model was lobbying ~ medicare spending, rather than medicare spending ~ lobbying). Python is not my strong suit, so I may be interpreting this wrong. Will merge this work in to keep building on it though.
adding my analysis where I try to find the relationship between pharmaceutical company's medicare income and lobbying expenditure. initially I was trying to do a time series analysis, but only had a few years of time series data to work with, so pivoted to look at the relationship above.
I could not find at this time a strong correlation between medicare income and lobbying expenditure, however.