Using Big Data Analytics to Determine the Most Significant Attributes That Affect the Chances of Getting a Loan Approved
Session Number
D03
Advisor(s)
Charles Downing, Northern Illinois University
Location
B-108
Start Date
28-4-2016 8:00 AM
End Date
28-4-2016 8:25 AM
Apr 28th, 8:00 AM
Apr 28th, 8:25 AM
Using Big Data Analytics to Determine the Most Significant Attributes That Affect the Chances of Getting a Loan Approved
B-108
Comments
Hundreds of loan applications are being submitted in every state. Massachusetts is no exception. The purpose of this investigation is to find out what are the main attributes that increase or decrease the chance of a particular loan application to get approved in Massachusetts. After getting access to all of the loan data available in that state in 2010, courtesy to EMC Corporation, a logistic regression was ran on that dataset. Not surprisingly, it was found that having a higher annual income and median family income and asking for a lower loan amount both significantly increased one’s chance of getting a particular loan approved. It was also found that having the lender in the first lien position provided a much higher chance of approval than having the lender be in the subordinate lien position. The type and purpose of a particular loan requested also played a crucial role in the probability of one getting a loan approved. Lastly, race was found to play a minor, but significant role as it was also discovered that African Americans, Hispanics/Latinos, Asians, Hawaiians/ Pacific Islanders, Alaskan Natives/American Indians all had slightly lower chances of approval than Whites had.