Skip to content

4.2-final-2007

Final Exam, 2007 · Section 4.2: Projections onto Lines and Subspaces Source: finalexam_1.pdf, p. 172

Problem

Problem 3 (10 points) Let \(L\) be the line in \(R^3\) spanned by the vector \((1, 1, 1)^T\). Let \(P\) be the projection matrix for the projection onto the line \(L\).

(a) What are the eigenvalues of the matrix \(P\)? (Indicate their multiplicities.)

(b) Find an orthonormal basis of the orthogonal complement \(L^\perp\) to the line \(L\).

(c) Calculate the projection of the vector \((1, 2, 3)^T\) onto the line \(L\).

(d) Calculate the projection of the vector \((1, 2, 3)^T\) onto the orthogonal complement \(L^\perp\).

Solution (a) \(P\) is a projection matrix onto a subspace of dimension 1, so the eigenvalues are 1, 0, 0. (b) \[ \begin{bmatrix} 1/\sqrt{2} \\ 0 \\ -1/\sqrt{2} \end{bmatrix}, \quad \begin{bmatrix} 1/\sqrt{6} \\ -2/\sqrt{6} \\ 1/\sqrt{6} \end{bmatrix}. \] (c) \(p = \begin{bmatrix} 2 \\ 2 \\ 2 \end{bmatrix}\). (d) The projection onto \(L^\perp\) is \(b - p = \begin{bmatrix} -1 \\ 0 \\ 1 \end{bmatrix}\).

Keywords: projection, eigenvalues, orthonormal basis, orthogonal complement