Shrinkage in linear discriminant function analysis can be minimalized by
A)having the number of participants in the smallest group be 10 times the number of predictor variables
B)having the number of participants in the largest group be 10 times the number of predictor variables
C)having the number of participants in the smallest group be at least 30
D)having the number of participants in both groups be at least 150

Question

Quizplus · Accepted Answer

Shrinkage in linear discriminant function analysis is minimized when the sample size is large relative to the number of predictor variables. Specifically, having the number of participants in the smallest group be 10 times the number of predictor variables helps ensure that the model is not overfitting to the sample data, thus reducing shrinkage.

Shrinkage in Linear Discriminant Function Analysis Can Be Minimalized by