273 reputation
19
bio website
location
age
visits member for 3 years, 9 months
seen Apr 4 '13 at 7:22

Jul
2
awarded  Curious
Jun
13
awarded  Popular Question
Nov
15
awarded  Popular Question
Feb
4
accepted solve $ y = (A+B^{-1})x $ for $x$
Jan
26
comment solve $ y = (A+B^{-1})x $ for $x$
I can just calculate $solve(BA+I, By)$, no?
Jan
26
comment solve $ y = (A+B^{-1})x $ for $x$
Or is there some reason that doesn't work?
Jan
26
comment solve $ y = (A+B^{-1})x $ for $x$
I think for my purposes, I can just solve $$(BA+I)x = By$$ $$x = (BA + I)^{-1}By$$ Give that as an answer, and I'll accept.
Jan
26
revised solve $ y = (A+B^{-1})x $ for $x$
added 40 characters in body
Jan
26
comment solve $ y = (A+B^{-1})x $ for $x$
Yes, sorry, that is indeed what I meant.
Jan
26
comment solve $ y = (A+B^{-1})x $ for $x$
That sounds like just the kind of thing I need, but I don't know what that is. Can you elaborate? Maybe I'm just dense.
Jan
26
revised solve $ y = (A+B^{-1})x $ for $x$
added 61 characters in body; edited title
Jan
26
asked solve $ y = (A+B^{-1})x $ for $x$
Jan
14
accepted 1st-order linear ODE with tridiagonal matrix. Efficient solutions?
Jan
10
asked 1st-order linear ODE with tridiagonal matrix. Efficient solutions?
Jan
10
comment Is there general formula for the exponential of a tridiagonal matrix?
I'm looking for a way to compute exp(A*t)*x_0 cheaply when A's a symmetric tridiagonal matrix. I think I may just have to eigen-decompose A and do it that way. Luckily I only have to decompose A once, and then it's O(n**2), which I guess is okay. Since you should be able to compute A*x_0 in O(n) steps since its tridiagonal, I was hoping for something better, but maybe that's not possible.
Jan
10
comment Is there general formula for the exponential of a tridiagonal matrix?
Did you ever find a solution to your problem?
Jul
10
comment Random samples from a normal distribution without explicitly constructing a covariance matrix
Yes, but the tricky part is that I need it with the covariance matrix C^-1 not C
Jul
9
asked Random samples from a normal distribution without explicitly constructing a covariance matrix
May
13
awarded  Commentator
May
13
comment How do I prove positive definiteness for a matrix difference?
The question has not been edited substantially.