Chapter 7 Predictor Selection
“Transformation occurs when there has been a learning lesson and you choose to create a better choice.”
-- Andrea Reibmayr
In the previous chapter, we have established the idea that predictive data analysis (PDA) is a kind of model building. In its three-step process, the first step is the predictor selection. This is because most prediction model does not use all the attributes of the data samples. An only a small amount of predictors are used in a model. Therefore before a model can be constructed or trained, it is necessary to select predictors1.
if you are familiar with association and PCA analyses, you can jump to chapter 8↩︎