Exercise 4. Running Regression Models
The Run Existing Regression Model function allows using prebuilt regression models to predict forest metrics.
Go to ALS Forest > Regression Analysis > Run Existing Regression Model to access the following dialog window.
- Import Model File: Specify a model file produced by Linear Regression, SVM, ANN, or Random Forest algorithm (check Save Regression Model in Exercise 3).
- Independent Variables: A list of independent variables to input into the predictive model. Refer to Exercise 2 for details.
- Plot Type: Choose the proper plot type according to the data. Set the Length for Square plots or Radius for Circular plots.
- Output Path: Specify an output directory to store the outputs, including regression models, model predictions (.tif file), and regression prediction reports (.html file).
Note: Input Independent Variables have to match the ones used for the model building. Otherwise, the tool will fail.
The output Regression Prediction Report contains model parameters, R, R-Square, RMSE, Probability Value, and the Significance of K-Fold Test Result.