Problem 3-06: Consider the four points (x,y):(2,3),(1.5,2.3),(2,3.5), and (1,1). (a) Use the normal equation to determine the best-fit linear model y^i=0+1x(i). (b) By squaring the x-values in order to generate a second feature, use the normal equation to determine the best-fit quadratic model y^i=0+1xi+2x(i)2. (c) For the models found in (a) and (b), determine which model is the better fit according to the objective function J()=n1(yh(X))T(yh(X)), where h(X)=X, and X is the data matrix with a column of 1 's. Will it always be the case that this model fits better, or does it depend on the dataset