Rank of the product of three matrices
Clash Royale CLAN TAG#URR8PPP
up vote
3
down vote
favorite
Assume that $AinmathbbR^ntimes nu$, $BinmathbbR^nutimes nu$ and $CinmathbbR^nutimes n$. I'm interested in the rank of the matrix
$$AB^-1C.$$
$A$ and $C$ are full rank ($n$), and of course, $B$ is as well ($nu$). In light of this operation I'm wondering if:
- Is there any explicit condition so that $operatornamerank(M)=n$?
- Can I say something about $operatornamerank(AB^-1C)$ using $operatornamerank(ABC)$ instead? (that would be much easier for my problem).
linear-algebra matrix-rank
add a comment |Â
up vote
3
down vote
favorite
Assume that $AinmathbbR^ntimes nu$, $BinmathbbR^nutimes nu$ and $CinmathbbR^nutimes n$. I'm interested in the rank of the matrix
$$AB^-1C.$$
$A$ and $C$ are full rank ($n$), and of course, $B$ is as well ($nu$). In light of this operation I'm wondering if:
- Is there any explicit condition so that $operatornamerank(M)=n$?
- Can I say something about $operatornamerank(AB^-1C)$ using $operatornamerank(ABC)$ instead? (that would be much easier for my problem).
linear-algebra matrix-rank
add a comment |Â
up vote
3
down vote
favorite
up vote
3
down vote
favorite
Assume that $AinmathbbR^ntimes nu$, $BinmathbbR^nutimes nu$ and $CinmathbbR^nutimes n$. I'm interested in the rank of the matrix
$$AB^-1C.$$
$A$ and $C$ are full rank ($n$), and of course, $B$ is as well ($nu$). In light of this operation I'm wondering if:
- Is there any explicit condition so that $operatornamerank(M)=n$?
- Can I say something about $operatornamerank(AB^-1C)$ using $operatornamerank(ABC)$ instead? (that would be much easier for my problem).
linear-algebra matrix-rank
Assume that $AinmathbbR^ntimes nu$, $BinmathbbR^nutimes nu$ and $CinmathbbR^nutimes n$. I'm interested in the rank of the matrix
$$AB^-1C.$$
$A$ and $C$ are full rank ($n$), and of course, $B$ is as well ($nu$). In light of this operation I'm wondering if:
- Is there any explicit condition so that $operatornamerank(M)=n$?
- Can I say something about $operatornamerank(AB^-1C)$ using $operatornamerank(ABC)$ instead? (that would be much easier for my problem).
linear-algebra matrix-rank
edited Jul 19 at 15:32
user7530
33.4k558109
33.4k558109
asked Jul 19 at 15:15


Nico F.
1379
1379
add a comment |Â
add a comment |Â
1 Answer
1
active
oldest
votes
up vote
2
down vote
accepted
Let me answer you second question first: unfortunately no, the conditions are not equivalent. Take
$$A=beginbmatrix0 \ 1 \ 0endbmatrix, B = beginbmatrix0 & 1 & 0\0 & 0 & 1\ 1 & 0 & 0endbmatrix, C = beginbmatrix1 \ 0 \ 0endbmatrix.$$
Then $A^TBC = 0$ but $A^T B^-1C = 1.$ Here the matrix $B$ was carefully crafted to rotate the image of $C$ into the kernel of $A^T$, but not vice-versa.
For the first question, I know of no simple condition (simpler than performing the multiplication and probing the product's rank). You can take the SVD of $A$ and $C$ and try to reason how B maps the different subspaces to each other, but at that point you're already better off taking the determinant...
Thanks a million for the answer. Extremely clear example! I'll follow your advice with the SVD... It will be a little messy since $A$,$B$ and $C$ are the result of several Kronecker products/sums. Thanks again!
– Nico F.
Jul 20 at 9:28
add a comment |Â
1 Answer
1
active
oldest
votes
1 Answer
1
active
oldest
votes
active
oldest
votes
active
oldest
votes
up vote
2
down vote
accepted
Let me answer you second question first: unfortunately no, the conditions are not equivalent. Take
$$A=beginbmatrix0 \ 1 \ 0endbmatrix, B = beginbmatrix0 & 1 & 0\0 & 0 & 1\ 1 & 0 & 0endbmatrix, C = beginbmatrix1 \ 0 \ 0endbmatrix.$$
Then $A^TBC = 0$ but $A^T B^-1C = 1.$ Here the matrix $B$ was carefully crafted to rotate the image of $C$ into the kernel of $A^T$, but not vice-versa.
For the first question, I know of no simple condition (simpler than performing the multiplication and probing the product's rank). You can take the SVD of $A$ and $C$ and try to reason how B maps the different subspaces to each other, but at that point you're already better off taking the determinant...
Thanks a million for the answer. Extremely clear example! I'll follow your advice with the SVD... It will be a little messy since $A$,$B$ and $C$ are the result of several Kronecker products/sums. Thanks again!
– Nico F.
Jul 20 at 9:28
add a comment |Â
up vote
2
down vote
accepted
Let me answer you second question first: unfortunately no, the conditions are not equivalent. Take
$$A=beginbmatrix0 \ 1 \ 0endbmatrix, B = beginbmatrix0 & 1 & 0\0 & 0 & 1\ 1 & 0 & 0endbmatrix, C = beginbmatrix1 \ 0 \ 0endbmatrix.$$
Then $A^TBC = 0$ but $A^T B^-1C = 1.$ Here the matrix $B$ was carefully crafted to rotate the image of $C$ into the kernel of $A^T$, but not vice-versa.
For the first question, I know of no simple condition (simpler than performing the multiplication and probing the product's rank). You can take the SVD of $A$ and $C$ and try to reason how B maps the different subspaces to each other, but at that point you're already better off taking the determinant...
Thanks a million for the answer. Extremely clear example! I'll follow your advice with the SVD... It will be a little messy since $A$,$B$ and $C$ are the result of several Kronecker products/sums. Thanks again!
– Nico F.
Jul 20 at 9:28
add a comment |Â
up vote
2
down vote
accepted
up vote
2
down vote
accepted
Let me answer you second question first: unfortunately no, the conditions are not equivalent. Take
$$A=beginbmatrix0 \ 1 \ 0endbmatrix, B = beginbmatrix0 & 1 & 0\0 & 0 & 1\ 1 & 0 & 0endbmatrix, C = beginbmatrix1 \ 0 \ 0endbmatrix.$$
Then $A^TBC = 0$ but $A^T B^-1C = 1.$ Here the matrix $B$ was carefully crafted to rotate the image of $C$ into the kernel of $A^T$, but not vice-versa.
For the first question, I know of no simple condition (simpler than performing the multiplication and probing the product's rank). You can take the SVD of $A$ and $C$ and try to reason how B maps the different subspaces to each other, but at that point you're already better off taking the determinant...
Let me answer you second question first: unfortunately no, the conditions are not equivalent. Take
$$A=beginbmatrix0 \ 1 \ 0endbmatrix, B = beginbmatrix0 & 1 & 0\0 & 0 & 1\ 1 & 0 & 0endbmatrix, C = beginbmatrix1 \ 0 \ 0endbmatrix.$$
Then $A^TBC = 0$ but $A^T B^-1C = 1.$ Here the matrix $B$ was carefully crafted to rotate the image of $C$ into the kernel of $A^T$, but not vice-versa.
For the first question, I know of no simple condition (simpler than performing the multiplication and probing the product's rank). You can take the SVD of $A$ and $C$ and try to reason how B maps the different subspaces to each other, but at that point you're already better off taking the determinant...
answered Jul 19 at 15:47
user7530
33.4k558109
33.4k558109
Thanks a million for the answer. Extremely clear example! I'll follow your advice with the SVD... It will be a little messy since $A$,$B$ and $C$ are the result of several Kronecker products/sums. Thanks again!
– Nico F.
Jul 20 at 9:28
add a comment |Â
Thanks a million for the answer. Extremely clear example! I'll follow your advice with the SVD... It will be a little messy since $A$,$B$ and $C$ are the result of several Kronecker products/sums. Thanks again!
– Nico F.
Jul 20 at 9:28
Thanks a million for the answer. Extremely clear example! I'll follow your advice with the SVD... It will be a little messy since $A$,$B$ and $C$ are the result of several Kronecker products/sums. Thanks again!
– Nico F.
Jul 20 at 9:28
Thanks a million for the answer. Extremely clear example! I'll follow your advice with the SVD... It will be a little messy since $A$,$B$ and $C$ are the result of several Kronecker products/sums. Thanks again!
– Nico F.
Jul 20 at 9:28
add a comment |Â
Sign up or log in
StackExchange.ready(function ()
StackExchange.helpers.onClickDraftSave('#login-link');
);
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
StackExchange.ready(
function ()
StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2fmath.stackexchange.com%2fquestions%2f2856726%2frank-of-the-product-of-three-matrices%23new-answer', 'question_page');
);
Post as a guest
Sign up or log in
StackExchange.ready(function ()
StackExchange.helpers.onClickDraftSave('#login-link');
);
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Sign up or log in
StackExchange.ready(function ()
StackExchange.helpers.onClickDraftSave('#login-link');
);
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Sign up or log in
StackExchange.ready(function ()
StackExchange.helpers.onClickDraftSave('#login-link');
);
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Sign up using Google
Sign up using Facebook
Sign up using Email and Password