Re: eigenvector
Posted by
Art Kendall on
Dec 08, 2013; 3:33pm
URL: http://spssx-discussion.165.s1.nabble.com/eigenvector-tp5723495p5723504.html
in this situation the
"direction" of an eigenvector is purely arbitrary. Both
packages show the same solution except for the arbitrary
direction.
Are you trying to learn matrix algebra or are you trying to
implement some procedure.
If it is the latter what are you trying to do? it is possible that
there is already some procedure implemented either in SPSS or by
someone else using SPSS.
Art Kendall
Social Research Consultants
On 12/7/2013 7:39 AM, asghar [via SPSSX Discussion] wrote:
Hello to all
I have below correlation matrix. I got eigenvalues and
eigenvectors by using CALL EIGEN command in spss syntax. elements
of eigenvector for first eigenvalue are negative. I did this
analysis in SAS program. But elements of eigenvector for the first
einenvalue are positive. could anyone please help to find the
reason of this different results in two these softwares(SPSS and
SAS). Please tell me, which is correct solution.
with best regards,
asghar
R={1.00 , .407, .315, .410 , .348;
.407 , 1.00 , .151, .220, .220;
.315, .151, 1.00, .400, .077 ;
.410, .220, .400, 1.00, .160;
.348, .220, .077, .160, 1.00}.
SPSS
Run MATRIX procedure:
A
-.5519753361 -.1324885342 -.0380180912 -.1157827542
-.8141983758
-.4189906133 -.3527569116 -.7427286539 .1498984098
.3548155024
-.4076001635 .5931608825 .1222964640 .6790093781
.0775378203
-.4802953099 .3906879012 .1154858899 -.6909341149
.3548977842
-.3506402948 -.5945859744 .6470049115 .1602533429
.2814647639
B
2.119374527
1.026117465
.770116074
.590769943
.493621991
The SAS System
14:27 Thursday, December 7, 2013 13
A
2.1193745
1.0261175
0.7701161
0.5907699
0.493622
B
0.5519753 0.1324885 -0.038018 -0.115783
-0.814198
0.4189906 0.3527569 -0.742729 0.1498984
0.3548155
0.4076002 -0.593161 0.1222965 0.6790094
0.0775378
0.4802953 -0.390688 0.1154859 -0.690934
0.3548978
0.3506403 0.594586 0.6470049 0.1602533
0.2814648
Art Kendall
Social Research Consultants