Finding a least-squares regression line is a very useful tool in statistics. Usually this is done using several messy formulas, but we can use what we learned about projections to find the answer much more elegantly. A group of students took a quiz, and we want to use a sample of the results to predict other scores. Before the quiz, a sample of five students had missed 0,1,2, 3, and 6 classes. They scored, respectively, 9,8, 5, 6, and 2 points. Assume that there is an approximately linear relationship between the variables. 1. Let x be the number of classes a student misses (the explanatory variable) and y be the quiz score (the response variable.) Suppose all of these points lay on the same line, y = C + Dx. Write a system of (five) equations in the variables c and D. Then rewrite the system as a matrix equation. 2. This system (clearly) has no solution. Find the least-squares solution by left-multiplying by the matrix A". Use the solution to write the line of best fit = C + Dx. (Notation note: in statistics, is used for this line that approximates the points.) 3. Suppose another student with 3 absences takes the quiz. What score does your model predict they will get