You are not logged in.
Pages: 1
Thanks Much!!
Sorry about that -
the main question is just
1) Is the matrix listed in post #5 DIAGONALIZABLE?
the other two related questions are:
2) Does the fact that this is the 'transistion' or 'state' matrix for an absorbing Markov Chain tell us anything about whether it IS or ISN"T Diagonalizable?
and
3) (again taking into consideration the context -- that this is a 'transistion' or 'state' matrix for an absorbing Markov Chain). Would there be any special significance in knowing that this matrix IS or ISN'T diagonalizable?
Hope this helps.
Thanks much!!
Hi Bobbym (or anyone else reading this). I had a question come up about this problem that I didn't know how to answer, and I was wondering if you could help. (Re: the transition or state matrix that you have in post #5) is this matrix Diagonalizable? Since we get two different results (depending on the order of multiplication) when multiplying this matrix by its transpose, we know that it can't be converted to a diagonal matrix by a unitary transform -- but I don't know whether this means the matrix is NOT Diagonalizable. What might it mean to say or show that it is or isn't Diagonalizable?
I'm way out of my field with this question. Thanks so much for your help!
Thanks for the book recommendations! is the 4th ed. better than the later editions of the Mizrahi and Sullivan? Amazon has third-parties selling used copies of the 9th edition for a buck.
I didn't have any experience with the methods you used in post #3 and #5. I sat down with them both in a morning, and made sense of each. I find them both kinda elegant. the method in post #8 I trust, but to my eyes right now, it looks UGLY.
I've been talking my friends through the first two methods, and they get them, but then ask: "can't you just do it with BINOMIALS, or nCr, or Pascal's triangle.....or something simple like that". I tell them that my sense of Probability is that it gets DIFFICULT QUICKLY, and that while this problem is SIMPLE TO POSE it's a little harder to crack than some High-School level Probability questions that might seem somewhat similar on the surface.
.....but here's my question: would you use any of these terms:
>> BINOMIALS, or nCr, or Pascal's triangle
in connection with what you've done in Post #8?
And.....is there any way to "talk around" the idea of what x represents/relates to in this formula?
Also, could you tell me which vol. of the Feller has the 24 pages I'm looking for?
Thanks, bobbym!
NPI = No Pun Intended.
NPI, but I can't make Heads or Tails of the formula. I see your key for r, p and n. I'm wondering if there are words to use in talking about "x"?
Where I'm stuck is really at the beginning. Is there a TEXT or WEBPAGE that explores how this formula from Feller is derived?
Thanks a lot!
FANTASTIC! Three different methods is GREAT. I get what's going on in post #3, and in post #5. I'm having a hard time making it through post #8. bobbym, could you either write a bit more about how this third method is working, or maybe LINK to a page (or suggest a TEXT) that might help explain the details of this third method.
I enjoyed the time I spent with the first two methods. I hadn't seen either of these techniques before I found this page while trying to answer a question similar to tongzilla's. I'm looking forward to understanding the post #8 method, but, for the moment, I'm stuck.
Thanks for your help!!
Pages: 1