1.)

The p-values in Table 3.4 test the null hypothesis which suggests that the various advertising mediums (TV, Radio, and Newspaper) have no significant effect on sales.

-TV: Due to the small p-value we can reject the null and suggest that TV advertising does affect sales.

-Radio: Due to the small p-value we can reject the null and suggest that Radio advertising does affect sales.

-Newspaper: Due to the large p-value we can accept the null and suggest that Radio advertising does not affect sales.

2.)

-KNN Classification is used when the dependent variable or the variable you are trying to predict is categorical. To make a prediction using this methodology, one must first find the K nearest neighbors of the new data vector, then analyze their classes and choose the majority class as a predicted label.

-KNN regression is used when the outcome variable is numerical. TO predict the value for a new point, we must again find the K nearest neighbors but now average their numerical response values to get the prediction.

3.)

a.) iii is the correct answer. It seems that is GPA is less than 3.5 college grads will have a higher salary. Only if a HS student has a GPA over 3.5 will they be likely to make more on average. iii, is the only one that satisfies this.

b.) $137,100

c.) False, we must consider the scale of the variables and magnitude of an effect depends on both the coefficients magnitude and its uncertainty values. Thus, such a thing cannot be concluded from the coefficient alone.

5.)