REGRESSION MODELING USING PRINCIPAL COMPONENTS

Shahar Boneh, Gonzalo R. Mendieta
1992 Conference on Applied Statistics in Agriculture  
In this paper we present a new stepwise method for selecting predictor variables in linear regression models and its application to agricultural data analysis. This method is an extension of principal component regression, and it consists of iteratively selecting original predictor variables one at a time from repeatedly selected subsets of principal components. The reasoning behind the method and its implementation are discussed, and an example of applying the method to agricultural data is
more » ... en. The example also demonstrates the advantages of the proposed method over some known methods.
doi:10.4148/2475-7772.1408 fatcat:6q47oqaqaffrjdokobfa4xf45q