Skip to content

Using ETL (Extract, Transform, Load) approach which serves greatly to extract data from data sources that are not optimized for analytics, move to a central host, and finally optimize it for analytics.

Notifications You must be signed in to change notification settings

mserobabina/Banking_deserts

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

21 Commits
 
 
 
 

Repository files navigation

ETL Project

Summary

The main objective of this analysis is to determine where bank deserts are located in the United States and populations affected by the lack of financial services in their area. Exploring who borrows, where they borrow and why. This analysis will be covering demographics and bank metrics for year 2017 (partial analysis of 2019 is presented as well) in 3,142 U.S. counties, excluding Puerto Rico.

Data Extraction

Three data sources were utilized to compile database:

  • For bank metrics, quarterly data from the FDIC Financial Data was utilized . CSV files can be extracted here
  • For demographics, data from the 2017 estimates American Community Survey (2010 U.S. Census) was utilized. CSV files can be extracted here
  • For unemployment and median household income, data from the Bureau of Labor Statistics - LAUS data was utilized. Excel files can be extracted here

Data Transformation

For each source:

  • Selected the variables of interest for the analysis
  • Eliminated any duplicates or null values
  • Summed quartely values to obtain annual observations for bank metrics

Note: For details on data transformation, please refer to the "Data_Transformation.ipynb"

Data Loading

Financial Deserts database were created in SQL, containing:

  • Demographics data information, using FIPS as primary key
  • Income and unemployment, using FIPS and State as primary keys
  • Bank metrics

Statistical Summary

Test Image 3

About

Using ETL (Extract, Transform, Load) approach which serves greatly to extract data from data sources that are not optimized for analytics, move to a central host, and finally optimize it for analytics.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published