Kiva Project Overview

From Analytics Practicum
Jump to navigation Jump to search


 

Home

Project Overview

 

Project Findings

 

Project Management

 

Documentation

 

About Us

 

ANLY482 Main Page

Previous Current

Project Objectives

As the bulk of loan records are from the Philippines, this project covers the use of geospatial analysis and statistical techniques, specifically Kernel Density Analysis and Exploratory Spatial Data Analytics techniques aimed at studying how geographical locations affect the presence and concentration of loans, and how loans are dispersed across geography based on different industry sectors, and how the spatial patterns differ across the different cities and municipalities within the Visayas.

Data

There are 4 main data files we received for our exploration and analysis. The primary file we used for analysis is kiva_loans.csv, which contains the main important variables of each loan, such as:

  1. Funded amount of the loan
  2. Loan amount of the loan
  3. Sector which the loan is used for, such as agriculture, education
  4. Activity which the loan is being used to fund
  5. Country and Region where the loan is being used in
  6. Currency which the loan is being disbursed in
  7. Time which the loan was posted, funded and disbursed
  8. The term/duration of the loan in months before repayment
  9. Tags associated with the loan
  10. The repayment interval type, such as whether repayment was done weekly, monthly, irregularly or in bullet

The remaining files loan_theme_ids, loan_themes_by_region and kiva_mpi_region_locations provide secondary information. Those which are of use to us include:

  1. World region/continent which the country resides in
  2. Latitude and longitude of the region (we are using the GADM map to obtain more in-depth geographical information, and obtain a more precise latitude and longitude)
  3. Loan theme type of the loan
  4. Percentage of borrowers that are in rural areas for particular field partners