sfbrigade / datasci-sba

Solving problems with the Small Business Administration
10 stars 18 forks source link

Next Steps (2017-05-31) #5

Open VincentLa opened 7 years ago

VincentLa commented 7 years ago

From analysis perspective:

  1. What are different perspectives of analysis that we can put on this dataset that will help highlight areas that are in need of closer attention by the SBA?
  2. Failure rate of loans by geographic area?
  3. Failure rate of loans by lender/geographic area?

Identifying successful businesses?

  1. In a perfect world, given a portfolio of SBA participants, identify businesses that have been really successful and are generating really good progress.
  2. At the end of the day, the people running the businesses and their stories are what keeps the program funded.
  3. If Noah goes to an area he would really want to know which businesses in that area have really thrived and who he can talk about as success stories. --https://public.tableau.com/profile/zlatan.kremonic#!/vizhome/SBALoans_1/ZIP_Analysis --Business Analysis tab more filters? Industries? Employees?

What was the data generation process to get to the sfdo_clean data set?

  1. Original dataset comes from here: https://www.sba.gov/about-sba/sba-performance/open-government/foia/frequently-requested-records/sba-7a-504-loan-data-reports subsetted geographically, concatenated the 504/7a datasets, and then put into OpenRefine which is Google's data cleaning tool to clean up city names. Release to public should probably keep with just the national data set.