If you find any mistakes, please make a comment! Thank you.

Solution to Linear Algebra Hoffman & Kunze Chapter 6.3


Exercise 6.3.1

Solution: The minimal polynomial for the identity operator is $x-1$. It annihilates the identity operator and the monic zero degree polynomial $p(x)=1$ does not, so it must be the minimal polynomial. The minimal polynomial for the zero operator is $x$. It is a monic polynomial that annihilates the zero operator and again the monic zero degree polynomial $p(x)=1$ does not, so it must be the minimal polynomial.


Exercise 6.3.2

Solution: The characteristic polynomial is
$$\left|\begin{array}{ccc} x & 0 & -c\\-1 & x & -b\\0 & -1 & x-a\end{array}\right|=\left|\begin{array}{ccc} x & 0 & -c\\-1 & 0 & x^2-ax-b\\0 & -1 & x-a\end{array}\right|$$$$=1\cdot\left|\begin{array}{cc}x & -c\\ -1 & x^2-ax-b\end{array}\right|=x^3-ax^2-bx-c.$$Now for any $r,s\in F$
$$A^2+rA+s=\left[\begin{array}{ccc}0 & c & ac\\0 & b & c+ba\\1 & a & b+a^2\end{array}\right] + \left[\begin{array}{ccc}0 & 0 & rc\\r& 0 &rb\\0 & r& ra\end{array}\right] + \left[\begin{array}{ccc}s & 0& 0\\0 & s & 0\\0 & 0 & s\end{array}\right]$$$$=\left[\begin{array}{ccc}s & c & ac+rc\\r & b +s& c+ba+br\\1 & a+r & b+a^2+ra+s\end{array}\right]\not=0.$$Thus $f(A)\not=0$ for all $f\in F[x]$ such that $\deg(F)=2$. Thus the minimum polynomial cannot have degree two, it must therefore have degree three. Since it divides $x^3-ax^2-bx-c$ it must equal $x^3-ax^2-bx-c$.


Exercise 6.3.3

Solution: The characteristic polynomial equals
$$\left|\begin{array}{cccc}x-1 & -1 & 0 & 0\\
1 &x+ 1 & 0 & 0\\
2 & 2 & x-2 & -1\\
-1 & -1 & 1 & x\end{array}\right|=\left|\begin{array}{cc}x-1&-1\\1&x+1\end{array}\right|\cdot\left|\begin{array}{cc}x-2&-1\\1&x\end{array}\right|,$$$$=x^2(x^2-2x+1)=x^2(x-1)^2.$$Here we used (5-20) page 158. The minimum polynomial is clearly not linear, thus the minimal polynomial is one of $x^2(x-1)^2$, $x^2(x-1)$, $x(x-1)^2$ or $x(x-1)$. We will plug $A$ in to the first three and show it is not zero. It will follow that the minimum polynomial must be $x^2(x-1)^2$.
$$A^2=
\left[\begin{array}{cccc}
0 & 0 & 0 & 0\\
0 & 0 & 0 & 0\\
-3 & -3 & 3 & 2\\
2 & 2 & -2 & -1
\end{array}\right]$$$$A-I=
\left[\begin{array}{cccc}
0 & 1 & 0 & 0\\
-1 & -2 & 0 & 0\\
-2 & -2 & 1 & 1\\
1 & 1 & -1 & -1
\end{array}\right]$$and
$$(A-I)^2=
\left[\begin{array}{cccc}
-1 & -2 & 0 & 0\\
2 & 3 & 0 & 0\\
1 & 1 & 0 & 0\\
0 & 0 & 0 & 0
\end{array}\right]$$Thus
$$A^2(A-I)=
\left[\begin{array}{cccc}
0 & 0 & 0 & 0\\
0 & 0 & 0 & 0\\
-1 & -1 & 1 & 1\\
1 & 1 & -1 & -1
\end{array}\right]\not=0$$$$A(A-I)^2=
\left[\begin{array}{cccc}
1 & 1 & 0 & 0\\
-1 & -1& 0 & 0\\
0 & 0 & 0 & 0\\
0& 0& 0 & 0
\end{array}\right]\not=0$$and
$$A(A-I)=
\left[\begin{array}{cccc}
-1 & -1 & 0 & 0\\
1 & 1& 0 & 0\\
-1 & -1 & 1 & 1\\
1& 1& -2 & -2
\end{array}\right]\not=0.$$Thus the minimal polynomial must be $x^2(x-1)^2$.


Exercise 6.3.4

Solution: Not diagonalizable, because for characteristic value $c=0$ the matrix $A-cI=A$ and $A$ is row equivalent to
$$\left[\begin{array}{cccc}1&1&0&0\\0&0&1&0\\0&0&0&1\\0&0&0&0\end{array}\right]$$which has rank three. So the null space has dimension one. So if $W$ is the null space for $A-cI$ then $W$ has dimension one, which is less than the power of $x$ in the characteristic polynomial. So by Theorem 2, page 187, $A$ is not diagonalizable.


Exercise 6.3.5

Solution: $T^k=0$ $\Rightarrow$ the only characteristic value is zero. We know the minimal polynomial divides this so the minimal polynomial is of the form $t^r$ for some $1\leq r\leq n$. Thus by Theorem 3, page 193, the characteristic polynomial's only root is zero, and the characteristic polynomial has degree $n$. So the characteristic polynomial equals $t^n$. By Theorem 4 (Caley-Hamilton) $T^n=0$.


Exercise 6.3.6

Solution: If $A^2=0$ and $A\not=0$ then the minimal polynomial is $x$ or $x^2$. So any $A\not=0$ such that $A^2=0$ has minimal polynomial $x^2$. E.g.
$$A=\left[\begin{array}{ccc}0&0&0\\1&0&0\\0&0&0\end{array}\right].$$


Exercise 6.3.7

Solution: $1,x,x^2,\dots,x^n$ is a basis.
$$1\mapsto0$$$$x\mapsto1$$$$x^2\mapsto2x$$$$\vdots$$$$\rule{4mm}{0mm}x^n\mapsto nx^{n-1}$$The matrix for $D$ is therefore
$$\left[\begin{array}{cccccc}
0 & 1 & 0 & 0 & \cdots & 0\\
0 & 0 & 2 & 0 & \cdots & 0\\
0 & 0 & 0 & 3 & \cdots & 0\\
\vdots & \vdots & \vdots & \vdots & \ddots & \vdots\\
0 & 0 & 0 & 0 & \cdots &n\\0 & 0 & 0 & 0 & \cdots &0\end{array}\right]$$Suppose $A$ is a matrix such that $a_{ij}=0$ except when $j=i+1$. Then $A^2$ has $a_{ij}=0$ except when $j=i+2$. $A^3$ has $a_{ij}=0$ except when $j=i+3$. Etc., where finally $A^{n+1}=0$. Thus if $a_{ij}\not=0$ $\forall$ $j=i+1$ then $A^k\not=0$ for $k<{n+1}$ and $A^{n+1}=0$. Thus the minimum polynomial divides $x^{n+1}$ and cannot be $x^k$ for $k<{n+1}$. Thus the minimum polynomial is $x^{n+1}$.


Exercise 6.3.8

Solution: $P$ can be given in the standard basis by left multiplication by $A=\left[\begin{array}{cc}1&0\\0&0\end{array}\right]$. Since $P$ is given by left multiplication by a matrix, $P$ is clearly linear. Since $A$ is diagonal, the characteristic values are the diagonal values. Thus the characteristic values of $A$ are $0$ and $1$. The characteristic polynomial is a degree two monic polynomial for which both $0$ and $1$ are roots. Therefore the characteristic polynomial is $x(x-1)$. If the characteristic polynomial is a product of distinct linear terms then it must equal the minimal polynomial. Thus the minimal polynomial is also $x(x-1)$.


Exercise 6.3.9

Solution: Suppose $A$ is $n\times n$. Claim: $|xI-A|=x^n+\text{trace}(A)x^{n-1}+\cdots$. Proof by induction: case $n=2$. $A=\left[\begin{array}{cc}a&b\\c&d\end{array}\right]$. $|xI-A|=x^2-(a+d)x+(ad-bc)$. The trace of $A$ is $a+d$ so we have established the claim for the case $n=2$. Suppose true for up to $n-1$. Let $r=a_{22}+a_{33}+\cdots+a_{nn}$. Then
$$\left|\begin{array}{cccc}
x-a_{11} & a_{12} & \cdots & a_{1n}\\
a_{21} & x-a_{22} & \cdots & a_{2n}\\
\ddots & \ddots & \cdots & \ddots\\
a_{n1} & a_{n2} & \cdots & x-a_{nn}
\end{array}\right|$$Now expanding by minors using the first column, and using induction, we get that this equals
$$(x-a_{11})(x^{n-1}-rx^{n-2}+\cdots)$$$$-a_{21}(\text{polynomial of degree $n-2$})$$$$+a_{31}(\text{polynomial of degree $n-2$})$$$$+\cdots$$$$=x^n+(r+a_{11})x^{n-1}+\text{polynomial of degree at most $n-2$}$$$$=x^n-\text{tr$(A)$}x^{n-1} +\cdots$$Now if $f(x)=(x-c_1)^{d_1}\cdots(x-c_k)^{d_k}$ then the coefficient of $x^{n-1}$ is $c_1d_1+\cdots c_kd_k$ so it must be that $c_1d_1+\cdots c_kd_k=\text{tr}(A)$.


Exercise 6.3.10

Solution: If we represent a $n\times n$ matrix as a column vector by stacking the columns of the matrix on top of each other, with the first column on the top, then the transformation $T$ is represented in the standard basis by the matrix
$$M=\left[\begin{array}{cccc}A & & & \\& A & \Large 0 & \\ & \Large 0 & \ddots & \\ & & & A\end{array}\right].$$And since
$$f(M)=\left[\begin{array}{cccc}f(A) & & & \\& f(A) & \Large 0 & \\ & \Large 0 & \ddots & \\ & & & f(A)\end{array}\right]$$it is evident that $f(M)=0$ $\Leftrightarrow$ $f(A)=0$.


Exercise 6.3.11

Solution: We have
\begin{align*}\begin{vmatrix}I_n & A\\ B & xI_n\end{vmatrix}=&\ \begin{vmatrix}I_n & A\\ B & xI_n\end{vmatrix}\begin{vmatrix}I_n & -A\\ 0 & I_n\end{vmatrix}\\=&\ \begin{vmatrix}I_n & 0\\ B & xI_n-BA\end{vmatrix}\\=&\ |xI_n-BA|\end{align*}Similarly,\begin{align*}\begin{vmatrix}I_n & A\\ B & xI_n\end{vmatrix}=&\ \begin{vmatrix}I_n & A\\ B & xI_n\end{vmatrix}\begin{vmatrix}I_n & 0\\ -Bx^{-1} & I_n\end{vmatrix}\\=&\ \begin{vmatrix}I_n-x^{-1}AB & A\\ 0 & xI_n\end{vmatrix}\\=&\ x^n|I_n-x^{-1}AB|=|xI_n-AB|.\end{align*}So $AB$ and $BA$ have the same characteristic polynomials.

But the minimum polynomials need not be equal. To see this let $A=\left[\begin{array}{cc}0&0\\1&0\end{array}\right]$ and $B=\left[\begin{array}{cc}1&0\\0&0\end{array}\right]$. Then $AB=\left[\begin{array}{cc}0&0\\1&0\end{array}\right]$ and $BA=\left[\begin{array}{cc}0&0\\0&0\end{array}\right]$ so the minimal polynomial of $BA$ is $x$ and the minimal polynomial of $AB$ is clearly not $x$ (it is in fact $x^2$).

Linearity

This website is supposed to help you study Linear Algebras. Please only read these solutions after thinking about the problems carefully. Do not just copy these solutions.

This Post Has 4 Comments

  1. In Exercise 6.3.11, why can we have 'the two polynomials are equal? '. I think we can't make sure that they both have n roots. (without using more properties of Field.)

    1. Yes, you are right. Here is the updated answer.

      We have
      \begin{align*}\begin{vmatrix}I_n & A\\ B & xI_n\end{vmatrix}=&\ \begin{vmatrix}I_n & A\\ B & xI_n\end{vmatrix}\begin{vmatrix}I_n & -A\\ 0 & I_n\end{vmatrix}\\=&\ \begin{vmatrix}I_n & 0\\ B & xI_n-BA\end{vmatrix}\\=&\ |xI_n-BA|\end{align*}Similarly,\begin{align*}\begin{vmatrix}I_n & A\\ B & xI_n\end{vmatrix}=&\ \begin{vmatrix}I_n & A\\ B & xI_n\end{vmatrix}\begin{vmatrix}I_n & 0\\ -Bx^{-1} & I_n\end{vmatrix}\\=&\ \begin{vmatrix}I_n-x^{-1}AB & A\\ 0 & xI_n\end{vmatrix}\\=&\ x^n|I_n-x^{-1}AB|=|xI_n-AB|.\end{align*}So $AB$ and $BA$ have the same characteristic polynomials.

  2. in question no. 7 it should be X^(n+1) rather than x^n

    1. Thank you, the size of the matrix is $(n+1)\times n+1$. I forget one more row.

Leave a Reply

Close Menu