6.2 Diagonalization of Matrices

Given an n×n matrix A, we may ask how many linearly independent eigenvectors the matrix A has. In Section 6.1, we saw several examples (with n=2 and n=3) in which the n×n matrix A has n linearly independent eigenvectors—the largest possible number. By contrast, in Example 5 of Section 6.1, we saw that the 2×2 matrix

A=[2302]

has the single eigenvalue λ=2 corresponding to the single eigenvector v=[10]T.

Something very nice happens when the n×n matrix A does have n linearly independent eigenvectors. Suppose that the eigenvalues λ1,λ2,,λn (not necessarily distinct) of A correspond to the n linearly independent eigenvectors v1,v2,,vn, respectively. Let

P=[|||v1v2vn|||]
(1)

be the n×n matrix having these eigenvectors as its column vectors. Then

AP=A[|||v1v2vn|||]=[|||Av2Av2Avn|||]

and hence

AP=[|||λ1v1λ2v2λnvn|||],
(2)

because Avj=λjvj for each j=1,2,,n. Thus the product matrix AP has column vectors λ1v1,λ2v2,,λnvn.

Now consider the diagonal matrix

D=[λ1000λ2000λn],
(3)

whose diagonal elements are the eigenvalues corresponding (in the same order) to the eigenvectors forming the columns of P. Then

PD=[|||v1v2vn|||][λ1000λ2000λn]=[|||λ1v2λ2v2λnvn|||],
(4)

because the product of the ith row of P and the jth column of D is simply the product of λj and the ith component of vj.

Finally, upon comparing the results in (2) and (4), we see that

AP=PD.
(5)

But the matrix P is invertible, because its n column vectors are linearly independent. So we may multiply on the right by P1 to obtain

A=PDP1.
(6)

Equation (6) expresses the n×n matrix A having n linearly independent eigenvectors in terms of the eigenvector matrix P and the diagonal eigenvalue matrix D. It can be rewritten as D=P1AP, but the form in (6) is the one that should be memorized.

Example 1

In Example 1 of Section 6.1 we saw that the matrix

A=[5622]

has eigenvalues λ1=2 and λ2=1 corresponding to the linearly independent eigenvectors v1=[21]T and v2=[32]T, respectively. Then

P=[2312],D=[2001],andP1=[2312].

So

PDP1=[2312][2001][2312]=[4322][2312]=[5622]=A,

in accord with Eq. (6).

Similarity and Diagonalization

The following definition embodies the precise relationship in (6) between the original matrix A and the diagonal matrix D.

Note that this relationship between A and B is symmetric, for if B=P1AP, then A=Q1BQ for some invertible matrix Q—just take Q=P1.

An n×n matrix A is called diagonalizable if it is similar to a diagonal matrix D; that is, there exist a diagonal matrix D and an invertible matrix P such that A=PDP1, and so

P1AP=D.
(8)

The process of finding the diagonalizing matrix P and the diagonal matrix D in (8) is called diagonalization of the matrix A. In Example 1 we showed that the matrices

A=[5622]andD=[2001]

are similar, and hence that the 2×2 matrix A is diagonalizable.

Now we ask under what conditions a given square matrix is diagonalizable. In deriving Eq. (6), we showed that if the n×n matrix A has n linearly independent eigenvectors, then A is diagonalizable. The converse of this statement is also true.

Remark

It is important to remember not only the fact that an n×n matrix A having n linearly independent eigenvectors is diagonalizable, but also the specific diagonalization A=PDP1 in Eq. (6), where the matrix P has the n eigenvectors as its columns, and the corresponding eigenvalues are the diagonal elements of the diagonal matrix D.

Example 2

In Example 5 of Section 6.1 we saw that the matrix

A=[2302]

has only one eigenvalue, λ=2, and that (to within a constant multiple) only the single eigenvector v=[10]T is associated with this eigenvalue. Thus the 2×2 matrix A does not have n=2 linearly independent eigenvectors. Hence Theorem 1 implies that A is not diagonalizable.

Example 3

In Example 6 of Section 6.1 we saw that the matrix

A=[30046216155]

has the following eigenvalues and associated eigenvectors:

λ1=3:v1=[102]Tλ1=1:v2=[025]Tλ1=0:v3=[013]T.

It is obvious (why?) that the three eigenvectors v1, v2, v3 are linearly independent, so Theorem 1 implies that the 3×3 matrix A is diagonalizable. In particular, the inverse of the eigenvector matrix

P=[v1v2v3]=[100021253]

is

P1=[100231452],

and the diagonal eigenvalue matrix is

D=[λ1000λ2000λ3]=[300010000].

Therefore, Eq. (6) in the form P1AP=D yields the diagonalization

P1AP=[100231452] [30046216155] [100021253]=[300010000]

of the matrix A.

The following theorem tells us that any set of eigenvectors associated with distinct eigenvalues (as in Example 3) is automatically linearly independent.

If the n×n matrix A has n distinct eigenvalues, then by Theorem 2 the n associated eigenvectors are linearly independent, so Theorem 1 implies that the matrix A is diagonalizable. Thus we have the sufficient condition for diagonalizability given in Theorem 3.

In general, however, an n×n matrix A can be expected to have fewer than n distinct eigenvalues λ1,λ2,,λk. If k<n, then we may attempt to diagonalize A by carrying out the following procedure.

  1. Step 1. Find a basis Si for the eigenspace associated with each eigenvalue λi.

  2. Step 2. Form the union S of the bases S1,S2,,Sk. According to Theorem 4 in this section, the set S of eigenvectors of A is linearly independent.

  3. Step 3. If S contains n eigenvectors v1,v2,,vn, then the matrix

    P=[v1v2vn]

    diagonalizes A: that is, P1AP=D, where the diagonal elements of D are the eigenvalues (repeated as necessary) corresponding to the n eigenvectors v1, v2, , vn.

If the set S—obtained by “merging” the bases for all the eigenspaces of A—contains fewer than n eigenvectors, then it can be proved that the matrix A is not diagonalizable.

Example 4

In Example 7 of Section 6.1, we saw that the matrix

A=[421201223]

has only two distinct eigenvalues, λ1=2 and λ2=3. We found that the eigenvalue λ1=2 corresponds to a 2-dimensional eigenspace with basis vectors v1=[110]T and v2=[102]T, and that λ2=3 corresponds to a 1-dimensional eigenspace with basis vector v3=[111]T. By Theorem 4 (or by explicit verification), these three eigenvectors are linearly independent, so Theorem 1 implies that the 3×3 matrix A is diagonalizable. The diagonalizing matrix

P=[v1v2v3]=[111101021]

has inverse matrix

P1=[231110221],

so we obtain the diagonalization

P1AP=[231110221] [421201223] [111101021]=[200020003]=D

of the matrix A.

6.2 Problems

In Problems 1 through 28, determine whether or not the given matrix A is diagonalizable. If it is, find a diagonalizing matrix P and a diagonal matrix D such that P1AP=D.

  1. [5421]

     

  2. [6644]

     

  3. [5320]

     

  4. [5432]

     

  5. [9865]

     

  6. [106127]

     

  7. [61023]

     

  8. [111568]

     

  9. [1413]

     

  10. [3111]

     

  11. [5191]

     

  12. [1191613]

     

  13. [130020002]

     

  14. [221221221]

     

  15. [331221001]

     

  16. [320010441]

     

  17. [783673222]

     

  18. [652432223]

     

  19. [111241441]

     

  20. [20061126150]

     

  21. [010120111]

     

  22. [221120571]

     

  23. [241351111]

     

  24. [321101112]

     

  25. [1020012000100001]

     

  26. [1001010100110002]

     

  27. [1100011000110002]

     

  28. [1101011100210002]

     

  29. Prove: If the matrices A and B are similar and the matrices B and C are similar, then the matrices A and C are similar.

  30. Suppose that the matrices A and B are similar and that n is a positive integer. Prove that the matrices An and Bn are similar.

  31. Suppose that the invertible matrices A and B are similar. Prove that their inverses A1 and B1 are also similar.

  32. Show that if the n×n matrices A and B are similar, then they have the same characteristic equation and therefore have the same eigenvalues.

  33. Suppose that the n×n matrices A and B are similar and that each has n real eigenvalues. Show that detA=detB and that Tr A=Tr B. See Problems 38 and 39 in Section 6.1.

  34. Consider the 2×2 matrix

    A=[abcd]

    and let Δ=(ad)2+4bc. Then show that

    1. A is diagonalizable if Δ>0;

    2. A is not diagonalizable if Δ<0;

    3. If Δ=0, then A may be diagonalizable or it may not be.

  35. Let A be a 3×3 matrix with three distinct eigenvalues. Tell how to construct six different invertible matrices P1,P2,,P6 and six different diagonal matrices D1,D2,,D6 such that PiDi(Pi)1=A for each i=1,2,,6.

  36. Prove: If the diagonalizable matrices A and B have the same eigenvalues (with the same multiplicities), then A and B are similar.

  37. Given: The diagonalizable matrix A. Show that the eigenvalues of A2 are the squares of the eigenvalues of A but that A and A2 have the same eigenvectors.

  38. Suppose that the n×n matrix A has n linearly independent eigenvectors associated with a single eigenvalue λ. Show that A is a diagonal matrix.

  39. Let λi be an eigenvalue of the n×n matrix A, and assume that the characteristic equation of A has only real solutions. The algebraic multiplicity of λi is the largest positive integer p(i) such that (λλi)p(i) is a factor of the characteristic polynomial |AλI|. The geometric multiplicity of λi is the dimension q(i) of the eigenspace associated with λi. It can be shown that p(i)q(i) for every eigenvalue λi. Taking this as already established, prove that the given matrix A is diagonalizable if and only if the geometric multiplicity of each eigenvalue is equal to its algebraic multiplicity.

..................Content has been hidden....................

You can't read the all page of ebook, please click here login for view all page.
Reset
18.224.62.160