Question: Lambda () for nominal variables What it measures: Proportional Reduction in Error (PRE) in predicting one nominal variable using another. Symmetric vs asymmetric: Symmetric treats
Lambda () for nominal variables What it measures: Proportional Reduction in Error (PRE) in predicting one nominal variable using another. Symmetric vs asymmetric: Symmetric treats both variables as predictors of each other. Asymmetric (Y|X) uses X to predict Y (directional). How it works: Without knowing X, predict the modal category of Y for everyone. Errors = total cases minus the largest column total. Call this E1. With X known, predict the modal Y separately within each category of X; sum errors across rows. Call this E2
Step by Step Solution
There are 3 Steps involved in it
Get step-by-step solutions from verified subject matter experts
