Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

From |
Austin Nichols <[email protected]> |

To |
[email protected] |

Subject |
Re: st: Problem computing eigenvalues, STATA vs. MATA |

Date |
Sun, 9 Jan 2011 10:42:14 -0500 |

Zachary Neal <[email protected]>: You are confusing vectors and values there, and not sorting by values after computing the eigensystem, but note that the only real difference is that a couple of vectors are negated (any scalar multiple of an eigenvector is also an eigenvector and the negation has the same length so they are essentially equivalent) and Mata is using quad precision to do the computations. matrix define A = 0,1,1,0,0 \ 1,0,0,1,0 \ 1,0,0,0,1 \ 0,1,0,0,0 \ 0,0,1,0,0 matrix symeigen vectors1 values1 = A matrix list values1 matrix list vectors1 mata values2 = 0 vectors2 = 0 symeigensystem(st_matrix("A"),vectors2,values2) v2=sort((values2\vectors2)',1)' v2[1,.] v2[2..6,.] end On Sun, Jan 9, 2011 at 8:08 AM, Zachary Neal <[email protected]> wrote: > I am trying to obtain the eigenvalues of a symmetric matrix. However, > I get different results depending on whether I use STATA or MATA. For > example: > > matrix define A = 0,1,1,0,0 \ 1,0,0,1,0 \ 1,0,0,0,1 \ 0,1,0,0,0 \ 0,0,1,0,0 > matrix symeigen eigenvalues1 eigenvectors1 = A > matrix list eigenvalues1 > > mata: A = 0,1,1,0,0 \ 1,0,0,1,0 \ 1,0,0,0,1 \ 0,1,0,0,0 \ 0,0,1,0,0 > mata: eigenvalues2 = 0 > mata: eigenvectors2 = 0 > mata: symeigensystem(A,eigenvalues2,eigenvectors2) > mata: eigenvalues2 > > In this case, eigenvalues1 does not equal eigenvalues2. I believe the > results yielded by STATA (i.e. eigenvalues1) are what I'm looking for. > > Why do these two sets of commands yield different results? What > commands are necessary in MATA to yield the same eigenvalues that are > given by STATA? > > Thank you > Zachary Neal * * For searches and help try: * http://www.stata.com/help.cgi?search * http://www.stata.com/support/statalist/faq * http://www.ats.ucla.edu/stat/stata/

**Follow-Ups**:**Re: st: Problem computing eigenvalues, STATA vs. MATA***From:*Zachary Neal <[email protected]>

**Re: st: Problem computing eigenvalues, STATA vs. MATA***From:*Steven Samuels <[email protected]>

**References**:**st: Problem computing eigenvalues, STATA vs. MATA***From:*Zachary Neal <[email protected]>

- Prev by Date:
**st: Problem computing eigenvalues, STATA vs. MATA** - Next by Date:
**st: generating variables for group members conditional on member's own values and values for other group members** - Previous by thread:
**st: Problem computing eigenvalues, STATA vs. MATA** - Next by thread:
**Re: st: Problem computing eigenvalues, STATA vs. MATA** - Index(es):