Solution

ID: past-exam-of-the-numerical-analysis-course-of-the-university-of-oxford/2014/2/b/solution

Write . The allowed orthogonal projection characterization, applied to the basis , gives the normal equations
Explicitly if is odd and if it is even. These equations can also be derived by differentiating with respect to each coefficient.
The Gram matrix is a symmetric positive-definite matrix. For any nonzero real coefficient vector ,
The polynomial inside the square is nonzero, so its square is positive on some interval. Hence has zero kernel and is invertible. Thus the normal equations have exactly one solution, giving exactly one best least-squares polynomial. This proves both existence and uniqueness. For completeness, if the residual is orthogonal to , then for every the Pythagorean theorem gives , directly verifying minimality and uniqueness.

New to topics? Read the docs here!