Model > Gradient Boosted Trees

Vincent R. Nijs, Rady School of Management (UCSD)

Estimate Gradient Boosted Trees

To estimate a Gradient Boosted Trees model model select the type (i.e., Classification or Regression), response variable, and one or more explanatory variables. Press the Estimate button or CTRL-enter (CMD-enter on mac) to generate results.

The model can be “tuned” by changing by adjusting the parameter inputs available in Radiant. In addition to these parameters, any others can be adjusted in Report > Rmd. The best way to determine the optimal values for all these hyper-parameters is to use Cross-Validation. In Radiant, you can use the cv.gbt function for this purpose. See the documentation for more information.

For more information on parameters that can be set for XGBoost, see the links below

• https://xgboost.readthedocs.io/en/latest/parameter.html
• https://xgboost.readthedocs.io/en/latest/tutorials/param_tuning.html

Report > Rmd

Add code to Report > Rmd to (re)create the analysis by clicking the icon on the bottom left of your screen or by pressing ALT-enter on your keyboard.

R-functions

For an overview of related R-functions used by Radiant to estimate a neural network model see Model > Neural network.

The key function from the xgboost package used in the gbt tool is xgboost.

© Vincent Nijs (2018)