To help get you started, here are some data exploration ideas:
See this forum thread for more ideas, and post there if you want to add your own ideas or answer some of the open questions!
This data was originally prepared and released by the Centers for Medicare & Medicaid Services (CMS). Please read the CMS Disclaimer-User Agreement before using this data.
Here, we've processed the data to facilitate analytics. This processed version has three components:
The original versions of the 2014, 2015, 2016 data are available in the "raw" directory of the download and "../input/raw" on Kaggle Scripts. Search for "dictionaries" on this page to find the data dictionaries describing the individual raw files.
In the top level directory of the download ("../input" on Kaggle Scripts), there are six CSV files that contain the combined at across all years:
Additionally, there are two CSV files that facilitate joining data across years:
The "database.sqlite" file contains tables corresponding to each of the processed CSV files.
The code to create the processed version of this data is available on GitHub.
