Question

a)would be important to include a variable that is not significant?

b)would be important to exclude a variable that is highly significant?

give examples and why

Answer 1

a) this is a wrong statement.

for example,suppose we want to predict the sales based on the invested budget on advertising medium say TV, Radio, Newspaper and product ID. if in final model we include product id which is not highly significant in predicting total sales we will get inaccurate results.

so we need to exclude non significant variables

b) this is a wrong statement

if we consider above example and exclude advertising budget invested on TV we will get inaccurate outcomes. So it is important to include significant variables