Orthogonal and symmetric maps: Symmetric maps
Orthonormal bases and symmetric maps
From the previously proven properties of symmetric matrices we now derive their diagonalizability.
Diagonalizability of symmetric maps
Let #V# be an inner product space of finite dimension and let #L:V\rightarrow V# be a symmetric linear map. Then there exists an orthonormal basis for #V# consisting of eigenvectors of #L#.
In terms of matrices the result can be formulated as follows.
Diagonalizability of symmetric matrices Let #n# be a natural number and #A# a symmetric #(n\times n)#-matrix. Then there is an orthogonal matrix #X# such that #X\, A\, X^{-1}# is a diagonal matrix.
#\alpha = # #\basis{\left[ {{1}\over{\sqrt{3}}} , -{{1}\over{\sqrt{3}}} , -{{1}\over{\sqrt{3}}} \right] , \left[ {{1}\over{\sqrt{2}}} , 0 , {{1}\over{\sqrt{2}}} \right] , \left[ -{{1}\over{\sqrt{6}}} , -{{\sqrt{2}}\over{\sqrt{3}}} , {{1}\over{\sqrt{6}}} \right] } #
The characteristic polynomial of #A# is \[-x^3+2 x^2+24 x\] This factors as follows into linear polynomials: \[p_A(x) = \left(6-x\right)\cdot x\cdot \left(x+4\right)\] Thus, the symmetric matrix #A# has eigenvalues #0#, #-4#, # 6#. Corresponding eigenvectors are
\[
E_{0}=\linspan{\left[ 1 , -1 , -1 \right] },\ E_{-4}=\linspan{\left[ 1 , 0 , 1 \right] },\ E_{6}=\linspan{\left[ -1 , -2 , 1 \right] }
\] These eigenvectors are perpendicular to each other. Therefore, the choice of a vector of length #1# in each of #E_{0}#, #E_{-4}#, #E_{6}# leads to an orthonormal basis of eigenvectors of #A#:
\[
\alpha = \basis{\left[ {{1}\over{\sqrt{3}}} , -{{1}\over{\sqrt{3}}} , -{{1}\over{\sqrt{3}}} \right] ,\ \left[ {{1}\over{\sqrt{2}}} , 0 , {{1}\over{\sqrt{2}}} \right] ,\ \left[ -{{1}\over{\sqrt{6}}} , -{{\sqrt{2}}\over{\sqrt{3}}} , {{1}\over{\sqrt{6}}} \right] }
\]
The characteristic polynomial of #A# is \[-x^3+2 x^2+24 x\] This factors as follows into linear polynomials: \[p_A(x) = \left(6-x\right)\cdot x\cdot \left(x+4\right)\] Thus, the symmetric matrix #A# has eigenvalues #0#, #-4#, # 6#. Corresponding eigenvectors are
\[
E_{0}=\linspan{\left[ 1 , -1 , -1 \right] },\ E_{-4}=\linspan{\left[ 1 , 0 , 1 \right] },\ E_{6}=\linspan{\left[ -1 , -2 , 1 \right] }
\] These eigenvectors are perpendicular to each other. Therefore, the choice of a vector of length #1# in each of #E_{0}#, #E_{-4}#, #E_{6}# leads to an orthonormal basis of eigenvectors of #A#:
\[
\alpha = \basis{\left[ {{1}\over{\sqrt{3}}} , -{{1}\over{\sqrt{3}}} , -{{1}\over{\sqrt{3}}} \right] ,\ \left[ {{1}\over{\sqrt{2}}} , 0 , {{1}\over{\sqrt{2}}} \right] ,\ \left[ -{{1}\over{\sqrt{6}}} , -{{\sqrt{2}}\over{\sqrt{3}}} , {{1}\over{\sqrt{6}}} \right] }
\]
We can write down the matrix of #(L_A)_\alpha# without calculation, because along the diagonal the eigenvalues will appear:
\[
(L_A)_\alpha =\matrix{0 & 0 & 0 \\ 0 & -4 & 0 \\ 0 & 0 & 6 \\ }
\] By way of verification of the answer, we compose the transition matrix #X# from #\alpha# to the standard basis and we check whether conjugation by #X# produces the diagonal matrix with #0#, #-4#, #6# on the diagonal:
\[\begin{array}{rcl} X\, A\, X^{-1} &=&X\, A\, X^{\top} \\ &=& \matrix{{{1}\over{\sqrt{3}}} & -{{1}\over{\sqrt{3}}} & -{{1}\over{\sqrt{3}}} \\ {{1}\over{\sqrt{2}}} & 0 & {{1}\over{\sqrt{2}}} \\ -{{1}\over{\sqrt{6}}} & -{{\sqrt{2}}\over{\sqrt{3}}} & {{1}\over{\sqrt{6}}} \\ }\, \matrix{-1 & 2 & -3 \\ 2 & 4 & -2 \\ -3 & -2 & -1 \\ }\, \matrix{{{1}\over{\sqrt{3}}} & -{{1}\over{\sqrt{3}}} & -{{1}\over{\sqrt{3}}} \\ {{1}\over{\sqrt{2}}} & 0 & {{1}\over{\sqrt{2}}} \\ -{{1}\over{\sqrt{6}}} & -{{\sqrt{2}}\over{\sqrt{3}}} & {{1}\over{\sqrt{6}}} \\ }^{\top} \\ & =& \matrix{0 & 0 & 0 \\ -2^{{{3}\over{2}}} & 0 & -2^{{{3}\over{2}}} \\ -\sqrt{6} & -2\cdot \sqrt{6} & \sqrt{6} \\ } \, \matrix{{{1}\over{\sqrt{3}}} & {{1}\over{\sqrt{2}}} & -{{1}\over{\sqrt{6}}} \\ -{{1}\over{\sqrt{3}}} & 0 & -{{\sqrt{2}}\over{\sqrt{3}}} \\ -{{1}\over{\sqrt{3}}} & {{1}\over{\sqrt{2}}} & {{1}\over{\sqrt{6}}} \\ }\\ &=&\matrix{0 & 0 & 0 \\ 0 & -4 & 0 \\ 0 & 0 & 6 \\ }\end{array}\]
\[
(L_A)_\alpha =\matrix{0 & 0 & 0 \\ 0 & -4 & 0 \\ 0 & 0 & 6 \\ }
\] By way of verification of the answer, we compose the transition matrix #X# from #\alpha# to the standard basis and we check whether conjugation by #X# produces the diagonal matrix with #0#, #-4#, #6# on the diagonal:
\[\begin{array}{rcl} X\, A\, X^{-1} &=&X\, A\, X^{\top} \\ &=& \matrix{{{1}\over{\sqrt{3}}} & -{{1}\over{\sqrt{3}}} & -{{1}\over{\sqrt{3}}} \\ {{1}\over{\sqrt{2}}} & 0 & {{1}\over{\sqrt{2}}} \\ -{{1}\over{\sqrt{6}}} & -{{\sqrt{2}}\over{\sqrt{3}}} & {{1}\over{\sqrt{6}}} \\ }\, \matrix{-1 & 2 & -3 \\ 2 & 4 & -2 \\ -3 & -2 & -1 \\ }\, \matrix{{{1}\over{\sqrt{3}}} & -{{1}\over{\sqrt{3}}} & -{{1}\over{\sqrt{3}}} \\ {{1}\over{\sqrt{2}}} & 0 & {{1}\over{\sqrt{2}}} \\ -{{1}\over{\sqrt{6}}} & -{{\sqrt{2}}\over{\sqrt{3}}} & {{1}\over{\sqrt{6}}} \\ }^{\top} \\ & =& \matrix{0 & 0 & 0 \\ -2^{{{3}\over{2}}} & 0 & -2^{{{3}\over{2}}} \\ -\sqrt{6} & -2\cdot \sqrt{6} & \sqrt{6} \\ } \, \matrix{{{1}\over{\sqrt{3}}} & {{1}\over{\sqrt{2}}} & -{{1}\over{\sqrt{6}}} \\ -{{1}\over{\sqrt{3}}} & 0 & -{{\sqrt{2}}\over{\sqrt{3}}} \\ -{{1}\over{\sqrt{3}}} & {{1}\over{\sqrt{2}}} & {{1}\over{\sqrt{6}}} \\ }\\ &=&\matrix{0 & 0 & 0 \\ 0 & -4 & 0 \\ 0 & 0 & 6 \\ }\end{array}\]
Unlock full access
Teacher access
Request a demo account. We will help you get started with our digital learning environment.
Student access
Is your university not a partner?
Get access to our courses via Pass Your Math independent of your university. See pricing and more.
Or visit omptest.org if jou are taking an OMPT exam.
Or visit omptest.org if jou are taking an OMPT exam.