Previous chapters have dealt with a number of regression type models, linear and multiple regression (Chapters 4 and 8), generalized linear models (Chapters 9 and 10), mixed-effects regression for longitudinal data (Chapters 11 and 12) and generalized additive and nonlinear models (Chapters 13 and 14). These parametric regression methods are widely used, but they may not give faithful data descriptions when the assumptions on which they are based are not met, or in the presence of higher order interactions among some of the explanatory variables.
KeywordsRoot Node Terminal Node Split Function Random Number Seed Daughter Node
Unable to display preview. Download preview PDF.