These questions are about non-linear regression. You are going to use the Abalone dataset. The Abalone dataset is easily accessible online in a csv format using Excel. Kaggle is an easy route.

The outcome variable to predict is V9, which records the number of rings in the abalone (the number of rings, in turn, is related to the abalones age)

Question 2: This question is a new question, independent from question 1

2 a) Select one predictor that seems suitable to predict V9 based on a second-degree polynomial. Answer the following questions:

- Justify why you selected that predictor.

- Was adding a squared term of that variable justified from a statistical point of view? (i.e., Is the second-degree polynomial on that variable better than a linear equation based on that variable?)

- If the second-degree polynomial is justified, write out its equation.

2 b) Is a third-degree polynomial better than a second-degree polynomial? JUSTIFY

2 c) Can you find another predictor that, once added to the second-degree polynomial, improves this polynomial? This question refers to a case where you are trying to add only a linear term of another predictor.