# Purchased mortgages data

A team of researchers wants to use data about a borrower and the location of a property to predict the amount of a mortgage. Variables include the income, race, and gender of the borrower as well as the census tract location of the property, and other information about the borrower and the type of property.

You can use this data to demonstrate TreeNet® Regression.

The second data set includes a small number of new data points for the same variables. You can use this data set to demonstrate Predict for TreeNet® Regression.

Worksheet column Description
Loan Amount The loan amount in whole dollars
Annual Income The borrower's annual income in whole dollars
Income Ratio The ratio of a borrower's debts to income
Front End Ratio The ratio of mortgage payments to income
Back End Ratio The ratio of debt payments to income
Number of Borrowers The number of borrowers
Age The borrower's age in years
Co-Borrower Age The co-borrower's age in years
Tract Minority Percent The Census tract's minority percentage
Tract Income The Census tract's median family income in whole dollars
Local Income The local area median income in whole dollars
Area Income The area's median family income in whole dollars
First Time Home Buyer Whether the borrower is a first-time home buyer: 1 = Yes or 0 = No
Occupancy Code The occupancy code: 1 = Principal residence or owner-occupied, 2 = Second home, or 3 = Investment property or rental
Self-Employed Whether the borrower is self-employed: 1 = Yes or 2 = No
Co-Borrower Race 4 The co-borrower's fourth race or national origin: 1 = American Indian or Alaskan Native, 2 = Asian, 3 = Black or African American, 4 = Native Hawaiian or other Pacific Islander, 5 = White, 7 = Information not provided, or 8 = No co-borrower
Co-Borrower Race 5 The co-borrower's fifth race or national origin: 1 = American Indian or Alaskan Native, 2 = Asian, 3 = Black or African American, 4 = Native Hawaiian or other Pacific Islander, 5 = White, 7 = Information not provided, or 8 = No co-borrower
Loan Purpose The loan purpose: 1 = Purchase, 2 = Refinancing, 3 = Second mortgage, 4 = New construction, or 5 = Rehabilitation
Gender The borrower's gender: 1 = Male, 2 = Female, or 3 = Information not provided
Number of Units The number of units on the property: 1–4
Ethnicity The borrower's ethnicity: 1 = Hispanic or Latino, 2 = Not Hispanic or Latino, or 3 = Information not provided
Co-Borrower Race 3 The co-borrower's third race or national origin: 1 = American Indian or Alaskan Native, 2 = Asian, 3 = Black or African American, 4 = Native Hawaiian or other Pacific Islander, 5 = White, 7 = Information not provided, or 8 = No co-borrower
Co-Borrower Gender The co-borrower's gender: 1 = Male, 2 = Female, 3 = Information not provided, or 4 = No co-borrower
Race 2 The borrower's additional race or national origin: 1 = American Indian or Alaskan Native, 2 = Asian, 3 = Black or African American, 4 = Native Hawaiian or other Pacific Islander, 5 = White, or 7 = Information not provided
Co-Borrower Ethnicity The co-borrower's ethnicity: 1 = Hispanic or Latino, 2 = Not Hispanic or Latino, 3 = Information not provided, or 4 = No co-borrower
Credit Score The borrower's credit score categorized by range: 1 = less than 620, 2 = 620 to less than 660, 3 = 660 to less than 700, 4 = 700 to less than 760, 5 = 760 or greater, or 9 = missing
Co-Borrower Credit Score The co-borrower's credit score categorized by range: 1 = less than 620, 2 = 620 to less than 660, 3 = 660 to less than 700, 4 = 700 to less than 760, 5 = 760 or greater, or 9 = missing or no co-borrower
Race The borrower's race or national origin: 1 = American Indian or Alaskan Native, 2 = Asian, 3 = Black or African American, 4 = Native Hawaiian or other Pacific Islander, 5 = White, or 7 = Information not provided
Co-Borrower Race 2 The co-borrower's second race or national origin: 1 = American Indian or Alaskan Native, 2 = Asian, 3 = Black or African American, 4 = Native Hawaiian or other Pacific Islander, 5 = White, 7 = Information not provided, or 8 = No co-borrower
Co-Borrower Race The co-borrower's race or national origin: 1 = American Indian or Alaskan Native, 2 = Asian, 3 = Black or African American, 4 = Native Hawaiian or other Pacific Islander, 5 = White, 7 = Information not provided, or 8 = No co-borrower
Property Type The property type: 1 = Single family detached, 2 = De minimus planned unit development, 3 = Single family attached, 4 = Two family, 5 = Townhouse, 6 = Low-rise condominium, 7 = Planned unit development, 8 = Duplex, 9 = Three family, 10 = Four family, 11 = Hi-rise condominium, or 12 = Manufactured home
Federal District The name of the federal home loan bank district
State Code A code that represents the state where the property is located
County Code A code that represents the county where the property is located
Core Based Statistical Area A code that represents the state, county, and census tract where the property is located. A missing value represents a state, county, and tract combination that is not located in a core based statistical area

## Reference

These data were adapted based on a public data set containing information on federal home loan bank mortgages. Original data from fhfa.gov.

By using this site you agree to the use of cookies for analytics and personalized content.  Read our policy