It seems to be, since $detA = 0$, If $det(A-kI)=0$ the kI part should be zero. So we can take k = 0 because I is nonzero. But I haven't any idea to take mathematical representation to prove it.
Can you help me. :-)
It seems to be, since $detA = 0$, If $det(A-kI)=0$ the kI part should be zero. So we can take k = 0 because I is nonzero. But I haven't any idea to take mathematical representation to prove it.
Can you help me. :-)
On
Just look the follwing arguments: for clear concept of it.
(1) if $ lemda$ is a eigen value of a matrix $A$ then we get a nonzero vector $v$ (known as the eigen vector) such that
$A*v=lemda*v$
Then it can be rewritten as $ (A- lemda* indentity matrix)*v=0$
Looking carefully you can see that it is a homogeneous system of equation. For non trivial solution ( since the trivial solution would not work for us as we have nonzero vector $v$ )of it we should have
$ det (A-lemda * identity matrix)=0$
(2) Also you should know that the product of the eigen values of a matrix is equal to the determinent of that matirx.
Since det(A) is zero so one of the eigen values of the matrix $A$ is zero.But this does not tell that all the eigen values is zero. Which can be seen by the early users counterexample.
Let $A=\begin{pmatrix} a & b \\ c & d \end{pmatrix}$. Then $\det(A)=\det(A-kI)=0$ is equivalent to $ad=bc=(a-k)(d-k)$. This yields $k(k-(a+d))=0$. We see that either $k=0$, or $k=a+d$. In the latter case, $k$ need not be zero. So we obtain counter examples.
For example, take $$ A=\begin{pmatrix} 1 & 1 \\ 1 & 1 \end{pmatrix}, $$ and $k=2$. Then $\det(A)=\det(A-2I)=0$.