Dear Statalisters,
I posted this on the general forum yesterday and got no response. I suppose this forum is more appropriate for this topic.
my goal is compute eigenvector centrality for a 471*471 matrix , my matrix is not symmetric, however, a published paper I follow states that
a simple example :
matrix A=(0,64, 7, 0, 0, 0\64,1,29,11,0,0\7,29,41,50,24,0\0,11,50,17,14,18\ 0,0,24,14,1,0\0,0,0,18,0,1)
mata
A = st_matrix("A")
X=.
L=.
symeigensystem(A, X, L)
matrix A
a finance paper that I model after states " Eigenvector centrality (EC) cannot always be applied to asymmetric adjacency matrices (Bonacich and Lloyd, 2001). For simplicity, we make the adjacency matrix symmetric by taking the maximum value of the upper and lower triangles,
it seems that if I take above matrix A and keep it non-symmetrical, symeigensystem still produces output, but I want to ask :
1. is the output for eigenvector centrality (EC) still okay to use or
2. do i have to use eigensystem( ) or something else for computing EC?
e.g. non-symmetrical,
modified A =matrix A=(0,64, 7, 0, 0, 0\64,1,29,11,0,0\0,0,41,50,24,0\0,11,50,17,14,18\ 0,0,24,14,1,0\0,0,0,18,0,1)
thanks !
Rochelle
I posted this on the general forum yesterday and got no response. I suppose this forum is more appropriate for this topic.
my goal is compute eigenvector centrality for a 471*471 matrix , my matrix is not symmetric, however, a published paper I follow states that
a simple example :
matrix A=(0,64, 7, 0, 0, 0\64,1,29,11,0,0\7,29,41,50,24,0\0,11,50,17,14,18\ 0,0,24,14,1,0\0,0,0,18,0,1)
mata
A = st_matrix("A")
X=.
L=.
symeigensystem(A, X, L)
matrix A
a finance paper that I model after states " Eigenvector centrality (EC) cannot always be applied to asymmetric adjacency matrices (Bonacich and Lloyd, 2001). For simplicity, we make the adjacency matrix symmetric by taking the maximum value of the upper and lower triangles,
it seems that if I take above matrix A and keep it non-symmetrical, symeigensystem still produces output, but I want to ask :
1. is the output for eigenvector centrality (EC) still okay to use or
2. do i have to use eigensystem( ) or something else for computing EC?
e.g. non-symmetrical,
modified A =matrix A=(0,64, 7, 0, 0, 0\64,1,29,11,0,0\0,0,41,50,24,0\0,11,50,17,14,18\ 0,0,24,14,1,0\0,0,0,18,0,1)
thanks !
Rochelle
