Skip to main content
john stark's user avatar
john stark's user avatar
john stark's user avatar
john stark
Unregistered
  • Member for 11 years, 1 month
  • Last seen more than 11 years ago
Loading…
Loading…
Loading…
Loading…
comment
expected matrix inverse of circulant plus diagonal matrix with chi-square variables
Cyclic = circulant indeed. Also, the matrix R is formed from some Fourier transform, so the eigenvalues converge to this transform as $N$ grows. See the OP for an update of the text.
revised
Loading…
revised
Loading…
comment
expected matrix inverse of circulant plus diagonal matrix with chi-square variables
yes, for each element. From Matlab, it appears as if the solution is $(R+\lambda I)^{-1}$ for some positive value $\lambda$. If true in general, the problem reduces into finding $\lambda$. And I agree, the large limit is not very helpful, but laid my complete problem down anyway.
Loading…
answered
Loading…
revised
equalizing diagonals of a matrix inverse
added 9 characters in body
Loading…
revised
Expected mean square error of an estimation problem
[Edit removed during grace period]
Loading…
comment
equalizing diagonals of a matrix inverse
1) The matrix $X_2$ is not known and is really what one must optimize for so that $G_2=diag_k(G_2). 2) no, I mean a positive semi-definite matrix.
revised
Loading…
awarded
awarded
Loading…
revised
equalizing diagonals of a matrix inverse
added 56 characters in body
Loading…
awarded
awarded