Example of prediction with TreeNet® Regression


This command is available with the Predictive Analytics Module. Click here for more information about how to activate the module.

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

The researcher can use the gradient boosted regression tree model to predict response values for new observations.


This example uses the dataset from Fit Model, but prediction is also available when you use Discover Key Predictors to create the model.

  1. Complete Example of Fit Model for TreeNet® Regression.
  2. Open the sample data set PurchasedMortgagesPredictions.MTW.
  3. Ensure that the worksheet that contains the prediction data is active and click the Predict button at the bottom of the regression tree results.
  4. From the drop-down list, select Enter columns of values.
  5. Enter the following values:
    Annual Income Annual Income
    Income Ratio Income Ratio
    Front End Ratio Front End Ratio
    Back End Ratio Back End Ratio
    Number of Borrowers Number of Borrowers
    Age Age
    Co-Borrower Age Co-Borrower Age
    Tract Minority Percent Tract Minority Percent
    Tract Income Tract Income
    Local Income Local Income
    Area Income Area Income
    First Time Home Buyer First Time Home Buyer
    Occupancy Code Occupancy Code
    Self-Employed Self-Employed
    Co-Borrower Race 4 Co-Borrower Race 4
    Co-Borrower Race 5 Co-Borrower Race 5
    Loan Purpose Loan Purpose
    Gender Gender
    Number of Units Number of Units
    Ethnicity Ethnicity
    Co-Borrower Race 3 Co-Borrower Race 3
    Co-Borrower Gender Co-Borrower Gender
    Race 2 Race 2
    Co-Borrower Ethnicity Co-Borrower Ethnicity
    Credit Score Credit Score
    Co-Borrower Credit Score Co-Borrower Credit Score
    Race Race
    Co-Borrower Race 2 Co-Borrower Race 2
    Co-Borrower Race Co-Borrower Race
    Property Type Property Type
    Federal District Federal District
    State Code State Code
    County Code County Code
    Core Based Statistical Area Core Based Statistical Area
  6. Click OK.

Interpret the results

Minitab uses the gradient boosted regression trees in the results to estimate the fit for the a set of prediction values. The researchers find the predicted loan amounts for various settings of the predictors.