Invertible Matrices: Why These Statements Are Not Correct

  • MHB
  • Thread starter Yankel
  • Start date
  • Tags
    Matrices
In summary, the statements are not correct because for the first one, $a\cdot (B^{-1}A^{-1}B)^t$ will always be invertible as long as $A$ and $B$ are invertible and $a\neq 0$. And for the second one, it is possible for $a\cdot (A+B)$ to not be invertible, and if it is, its inverse will not have the form $\frac{1}{a}\cdot (A^{-1}+B^{-1})$.
  • #1
Yankel
395
0
I have one more question, I have two matrices A and B, both squared and with the same order. And I have a scalar, a not equal to zero.

why are these statements not correct ?

1. If A and B are invertible, then
[tex]a\cdot (B^{-1}A^{-1}B)^{^{t}}[/tex]
is not necessarily invertible

2. If A and B are invertible, then
[tex]a\cdot (A+B)[/tex]
is not necessarily invertible, but if it is, it's inverse is the matrix
[tex]\frac{1}{a}\cdot (A^{-1}+B^{-1})[/tex]

Thanks...
 
Physics news on Phys.org
  • #2
Yankel said:
I have one more question, I have two matrices A and B, both squared and with the same order. And I have a scalar, a not equal to zero.

why are these statements not correct ?

1. If A and B are invertible, then
[tex]a\cdot (B^{-1}A^{-1}B)^{^{t}}[/tex]
is not necessarily invertible

2. If A and B are invertible, then
[tex]a\cdot (A+B)[/tex]
is not necessarily invertible, but if it is, it's inverse is the matrix
[tex]\frac{1}{a}\cdot (A^{-1}+B^{-1})[/tex]

Thanks...

For the first one, I'm going to assume WLOG that $A$ and $B$ are $n\times n$ matrices. if $A$ and $B$ are invertible, then $\det(A)\neq 0$ and $\det(B)\neq 0$. Furthermore, if $a\neq 0$ is a scalar, then
\[\det (a\cdot(B^{-1}A^{-1}B)^t)=a^n\det((B^{-1}A^{-1}B)^t) = a^n \det(B^{-1}A^{-1}B)=\frac{a^n\det(B)}{\det(B)\det(A)}=\frac{a^n}{\det(A)}.\]
This tells me that we'll always have $a\cdot(B^{-1}A^{-1}B)^t$ invertible provided that $A$ and $B$ are invertible with a scalar $a\neq 0$.

For the second one, it's true that if $A$ and $B$ are invertible, it may so happen that $a\cdot (A+B)$ isn't invertible. For instance, take $A=I$, $B=-I$ and $a=1$, where $I$ is the identity matrix. Then both matrices are invertible, but $A+B=O$, where $O$ is the zero matrix, which is clearly not invertible.

If $a\cdot(A+B)$ was invertible, then it wouldn't have that form for the inverse. We can see this is the case since
\[\left(a\cdot(A+B)\right)\left(\frac{1}{a}\cdot(A^{-1}+B^{-1})\right)=AB^{-1}+BA^{-1}+2I\neq I\]I hope this helps!
 

FAQ: Invertible Matrices: Why These Statements Are Not Correct

1. What is an invertible matrix?

An invertible matrix is a square matrix that has a unique solution for every set of linear equations. In other words, it is a matrix that can be reversed or "inverted" to retrieve the original matrix.

2. How do you determine if a matrix is invertible?

A matrix is invertible if its determinant is non-zero. This means that the matrix has a unique solution for every set of linear equations and can be reversed to retrieve the original matrix.

3. What are the properties of an invertible matrix?

An invertible matrix has several properties, including a non-zero determinant, a unique solution for every set of linear equations, and the ability to be reversed to retrieve the original matrix. It also has a unique inverse, which is a matrix that, when multiplied by the original matrix, results in the identity matrix.

4. Are all square matrices invertible?

No, not all square matrices are invertible. A square matrix must have a non-zero determinant in order to be invertible. If the determinant is zero, then the matrix is not invertible and is known as a singular matrix.

5. What are some real-world applications of invertible matrices?

Invertible matrices have many practical applications, including in computer graphics, cryptography, and engineering. They are also used in solving systems of linear equations and in data compression algorithms.

Similar threads

Replies
1
Views
981
Replies
16
Views
2K
Replies
5
Views
2K
Replies
9
Views
4K
Replies
1
Views
2K
Back
Top