Question

12. What are the different methods of entry for predictors in multiple regression?

Answer 1

There are three ways of choosing the predictors in multiple

regression:

Forward Selection: In this process variables are selected

based on their significance one by one. It starts with an

empty model and adds one variable at a time based on

the improvement of the model.

Backward Elimination: In this process, first we select all the

variables and then remove one at a time based on their

contribution into the model.

Stepwise Selection: In stepwise selection, at each step a

variable is either added or removed from the model.

It starts with an empty model, then first the most

significant variable is included into the model. Then with

the presence of this variable, we check the significance of

the all the models taking one of the other variables at a

time. Then if any of those models are significant, then we

choose the most significant one. After that in presence of

the second variable, we check the significance of the first

added variable. If it is significant, we carry on adding

another variable at a time, and if it is not significant, we

drop the variable from the model and go ahead.