Ex. 18.7
Ex. 18.7
Consider a linear regression problem where , and assume the rank of is . Let the SVD of ,where is nonsingular, and is with orthonormal columns.
(a) Show that there are infinitely many least-squares solutions all with zero residuals.
(b) Show that the ridge-regression estimate for can be written
(c) Show that when , the solution has residuals all equal to zero, and is unique in that it has the smallest Euclidean norm amongst all zero-residual solutions.
Soln. 18.7
(a) Since has rank , we know there exists such that . If has zero residuals, so does for any . Therefore there are infinitely many least-squares solutions all with zero residuals.
(b) This is the same as Ex. 18.4.
(c) Note that
so has zero residual.
Suppose that has zero residual for some , that is, . Since has zero residual, we know
Note that is nonsingular, so we have . Now consider the Euclidean norm of , we have
Since , we know that has the smallest Euclidean norm.