Let $m\times m$ matrix B has eigenvalues $\beta_1,\beta_2, ... \beta_n$ and eigenvectors $b_1, b_2, ... b_n$. I need to find eigenvalues and eigenvectors for operator $$R: Mat_{n\times m} \rightarrow Mat_{n\times m}, R(x) = XB^T. $$

As I understand $XB^T = \beta X$ where $X$ is a left eigenvector and $\beta$ is a left eigenvalue.

Then $(XB^T)^T=BX^T$ and operator $$R^T(X): Mat_{m\times n} \rightarrow Mat_{m\times n}, R^T(X)=BX^T $$ After that I said that $BX^T=\beta'X^T$ and by using properties of determinants $det|B^T-\beta|=det|B-\beta|^T=det|B-\beta|=det|B-\beta'| \Rightarrow \beta = \beta'$.

Also I know eigenvalues and eigenvectors for $R^T(X)=BX^T$, they equal: $\beta_i \ and \ b_i. $

After all these considerations I found that $X^T and\ X$ are orthogonal and their product commutes with a diagonal matrix of eigenvalues, but I don't know how to find the explicit form of X.

Could you help please?

Could you correct my solving please?


Your operator $R^T$ is much more complicated to study than your operator $R$.

For $R$, see


Here $R=I\otimes B$ if we stack the matrices row by row.

The eigenvalues of $R$ are $\beta_1,\cdots,\beta_m$, each $n$ times. If $b_1,\cdots,b_m$ are linearly independent, then $B$ is diagonalizable and $R$ also. It is easy to obtain the associated eigenvectors.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.