Example of prediction with Random Forests® Regression

Note

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

A team of researchers collects data from the sale of individual residential properties in Ames, Iowa. The researchers want to identify the variables that affect the sale price. Variables include the lot size and various features of the residential property.

  1. Complete Example of Random Forests® Regression.
  2. Open the sample data AmesHousingPredictions.mtw.
  3. Ensure that the worksheet that contains the prediction data is active and select Predict in the results.
  4. From the drop-down list, select Enter columns of values.
  5. Enter the following values:
    Lot Frontage Lot Frontage
    Lot Area Lot Area
    Veneer Area Veneer Area
    Basement SF 1 Basement SF 1
    Basement SF 2 Basement SF 2
    Basement Unf SF Basement Unf SF
    Total Basement SF Total Basement SF
    1st Floor SF 1st Floor SF
    2nd Floor SF 2nd Floor SF
    Low Quality SF Low Quality SF
    Living Area SF Living Area SF
    Garage Area SF Garage Area SF
    Wood Deck SF Wood Deck SF
    Open Porch SF Open Porch SF
    Enclosed Porch Enclosed Porch
    Season Porch Season Porch
    Screen Porch Screen Porch
    Pool Area Pool Area
    Misc Value Misc Value
    Year Built Year Built
    Year Remod/Add Year Remod/Add
    Basement Full Bath Basement Full Bath
    Basement Half Bath Basement Half Bath
    Full Bath Full Bath
    Half Bath Half Bath
    Bedroom Bedroom
    Kitchen Kitchen
    Total Rooms Total Rooms
    Fireplaces Fireplaces
    Garage Year Garage Year
    Garage Cars Garage Cars
    Month Sold Month Sold
    Year Sold Year Sold
    Type Type
    Zoning Zoning
    Street Street
    Alley Alley
    Lot Shape Lot Shape
    Land Contour Land Contour
    Utilities Utilities
    Configuration Configuration
    Slope Slope
    Neighborhood Neighborhood
    Condition 1 Condition 1
    Condition 2 Condition 2
    Quality Quality
    Condition Condition
    Roof Style Roof Style
    Roof Material Roof Material
    Exterior 1st Exterior 1st
    Exterior 2nd Exterior 2nd
    Veneer Type Veneer Type
    Exterior Quality Exterior Quality
    Exterior Condition Exterior Condition
    Foundation Foundation
    Basement Height Basement Height
    Basement Condition Basement Condition
    Basement Exposure Basement Exposure
    Basement Finish Type 1 Basement Finish Type 1
    Basement Finish Type 2 Basement Finish Type 2
    Heating Heating
    Heating Quality Heating Quality
    Central Air Central Air
    Electrical Electrical
    Kitchen Quality Kitchen Quality
    Function Function
    Fireplace Quality Fireplace Quality
    Garage Type Garage Type
    Garage Finish Garage Finish
    Garage Quality Garage Quality
    Garage Condition Garage Condition
    Paved Drive Paved Drive
    Pool Quality Pool Quality
    Fence Fence
    Misc Feature Misc Feature
    Sale Type Sale Type
    Sale Condition Sale Condition
  6. Click OK.

Interpret the results

Minitab uses the random forest regression trees in the results to estimate the fit for the set of prediction values. The researchers find the predicted sales prices for various settings of the predictors.
Fit
224796
88291
522279
480260
216826
112932
137328
190311
229939
229610
362637
174576
238485
256864