Eigenvalues with the QR and Lanczos methods

Thus, let us introduce a transformation \( \mathbf{S_{1}} \) which operates like $$ \mathbf{S_{1}} = \left( \begin{array}{cccc} \cos \theta & 0 & 0 & \sin \theta\\ 0 & 0 & 0 & 0 \\ 0 & 0 & 0 & 0 \\ \cos \theta & 0 & 0 & \cos \theta \end{array} \right) $$ Then the similarity transformation $$ \mathbf{S_{1}^{T} A S_{1}} = \mathbf{A'} = \left( \begin{array}{cccc} d'_{1} & e'_{1} & 0 & 0 \\ e'_{1} & d_{2} & e_{2} & 0 \\ 0 & e_{2} & d_{3} & e'{3} \\ 0 & 0 & e'_{3} & d'_{4} \end{array} \right) $$ produces a matrix where the primed elements in \( \mathbf{A'} \) have been changed by the transformation whereas the unprimed elements are unchanged.