Difference between revisions of "Kiva Project Overview"

From Analytics Practicum
Jump to navigation Jump to search
 
(One intermediate revision by the same user not shown)
Line 14: Line 14:
 
| style="background:none;" width="1%" |  
 
| style="background:none;" width="1%" |  
 
| style="padding:0.2em; font-size:100%; background-color:#008000; text-align:center; color:#F5F5F5" width="10%" |  
 
| style="padding:0.2em; font-size:100%; background-color:#008000; text-align:center; color:#F5F5F5" width="10%" |  
[[Kiva Project Findings| <font color="#f1cf0e" size=3 face="Century Gothic"><b>Project Findings</b></font>]]
+
[[Kiva_Project_Findings_Final| <font color="#f1cf0e" size=3 face="Century Gothic"><b>Project Findings</b></font>]]
  
 
| style="background:none;" width="1%" | &nbsp;
 
| style="background:none;" width="1%" | &nbsp;
Line 47: Line 47:
 
<div style="height: 1em"></div>
 
<div style="height: 1em"></div>
 
<div><font face="Arimo" size="4">
 
<div><font face="Arimo" size="4">
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.
+
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.  
 
 
<div style="background: #FFD700; line-height: 0.3em; border-left: #008000 solid 13px;"><div style="border-left: #FFFFFF solid 5px; padding:15px;"><font face ="Elephant" color= "black" size="3">Project Methodology</font></div></div>
 
<div style="height: 1em"></div>
 
<div><font face="Arimo" size="4">
 
Our team will attempt to use geospatial analysis to find out the how the characteristics of borrowing activities in different geographical locations differ from each other, and analyze how the different attributes of the loan vary across time for each geographical region. Geospatial analysis will allow us to build maps and make the relationships between the other attributes and geolocation data understandable and insightful. From there, we will be able to obtain more accurate trend analysis to our objectives, such as the duration of loan term and the repayment period.
 
</font></div>
 
 
<div style="height: 2em"></div>
 
<div style="height: 2em"></div>
 
 
<div style="background: #FFD700; line-height: 0.3em; border-left: #008000 solid 13px;"><div style="border-left: #FFFFFF solid 5px; padding:15px;"><font face ="Elephant" color= "black" size="3">Data</font></div></div>
 
<div style="background: #FFD700; line-height: 0.3em; border-left: #008000 solid 13px;"><div style="border-left: #FFFFFF solid 5px; padding:15px;"><font face ="Elephant" color= "black" size="3">Data</font></div></div>
 
<div style="height: 1em"></div>
 
<div style="height: 1em"></div>

Latest revision as of 16:38, 15 April 2018


 

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