$(4$ points$)$ Adapt the codes we used in Lecture $5$ for College GPA

prediction $($codeeviewsGPA$2.$prg$)$ to do a prediction using a model that

uses the college GPA $($colgpa$)$ on the left hand side and in addition to the

high school percentile ranking $($hsperc$)$ and SAT score $($sat$)$ variables we

used in class, it also uses the athlete status $($athlete$)$ and gender $($female$).$

Make a point prediction and a $95\%$ confidence interval for two

individuals:

a$.$ Person $1$: man, athlete, in the $40\%$ of high school percentile ranking,

and a SAT score equal to $1200.$

b$.$ Person $2$: woman, non$-$athlete, in the $15\%$ of high school percentile

ranking, and a SAT score equal to $1450.$

The code you write $($again$,$ adapting from what we've used in class$)$ should

read the raw data, run the regression and do the prediction, including the

confidence intervals using the formula on the slides. $($You can remove the

part of the code that produces the initial data plots.$)$ Upload the code to

ELMS and also copy the code to your write$-$up$.$ In your write$-$up report the

point estimate and the confidence interval you find.

$(2$ points$)$ Using the same data, answer the following question. You do not

need to write a code to do this, but you need show the regression output

you use to answer the question.

Controlling for high school percentile and gender, does being an athlete

change how one's SAT affects one's college GPA? $($In other words, would a

$1$ unit change in SAT have the same effect on college GPA for athlete and non$-$

athlete?$)$

'Second image is the code codeeviewsGPA$2.$prg