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]
st: Factors correlated after -predict-... What is going wrong?
From
Trevor Zink <[email protected]>
To
<[email protected]>
Subject
st: Factors correlated after -predict-... What is going wrong?
Date
Wed, 11 Dec 2013 10:55:33 -0800
Hi Statalist,
I am using -factor- to develop three factors, rotating them using
-rotate, varimax- and then produce variables from the factors using
-predict-.
Varimax is orthogonal rotation so should produce factors with zero
correlation. Testing the factors' correlation after rotation with -estat
common- produces the expected result, that correlation is 0. However,
after I produce variables from the factors using -predict-, these new
variables are correlated. How? Why? I tried replicating the steps using
the example dataset from the manual (/r12/sp2), and in that case the
predicted variables also have zero correlation. So, I guess it's
something unique to my dataset, but I have no idea what. Any ideas?
Thanks,
Trevor Zink
Here's a code/out snippet:
. factor q3b q3c q3d q3e q3f q3g q3h q3l q3m, factor(3)
(obs=112)
Factor analysis/correlation Number of obs
= 112
Method: principal factors Retained factors
= 3
Rotation: (unrotated) Number of params
= 24
--------------------------------------------------------------------------
Factor | Eigenvalue Difference Proportion Cumulative
-------------+------------------------------------------------------------
Factor1 | 2.05668 0.64145 0.5745 0.5745
Factor2 | 1.41523 0.84971 0.3953 0.9698
Factor3 | 0.56551 0.28624 0.1580 1.1278
Factor4 | 0.27927 0.17225 0.0780 1.2058
Factor5 | 0.10702 0.24115 0.0299 1.2357
Factor6 | -0.13413 0.07784 -0.0375 1.1982
Factor7 | -0.21198 0.02515 -0.0592 1.1390
Factor8 | -0.23712 0.02337 -0.0662 1.0728
Factor9 | -0.26049 . -0.0728 1.0000
--------------------------------------------------------------------------
LR test: independent vs. saturated: chi2(36) = 290.70 Prob>chi2 =
0.0000
Factor loadings (pattern matrix) and unique variances
-----------------------------------------------------------
Variable | Factor1 Factor2 Factor3 | Uniqueness
-------------+------------------------------+--------------
q3b | 0.5642 0.2899 -0.2814 | 0.5184
q3c | 0.4128 0.5956 -0.1803 | 0.4424
q3d | 0.1122 0.5641 -0.1312 | 0.6520
q3e | 0.2152 0.3567 0.0335 | 0.8253
q3f | 0.4873 -0.2166 0.0502 | 0.7131
q3g | 0.7222 -0.3959 -0.0625 | 0.3179
q3h | 0.7545 -0.4174 -0.0195 | 0.2561
q3l | 0.2966 0.3510 0.3937 | 0.6339
q3m | 0.3045 0.1735 0.5232 | 0.6034
-----------------------------------------------------------
. rotate, varimax blanks(.3) factor(3)
Factor analysis/correlation Number of obs
= 112
Method: principal factors Retained factors
= 3
Rotation: orthogonal varimax (Kaiser off) Number of params
= 24
--------------------------------------------------------------------------
Factor | Variance Difference Proportion Cumulative
-------------+------------------------------------------------------------
Factor1 | 1.86726 0.46529 0.5216 0.5216
Factor2 | 1.40196 0.63377 0.3916 0.9132
Factor3 | 0.76820 . 0.2146 1.1278
--------------------------------------------------------------------------
LR test: independent vs. saturated: chi2(36) = 290.70 Prob>chi2 =
0.0000
Rotated factor loadings (pattern matrix) and unique variances
-----------------------------------------------------------
Variable | Factor1 Factor2 Factor3 | Uniqueness
-------------+------------------------------+--------------
q3b | 0.3402 0.6048 | 0.5184
q3c | 0.7316 | 0.4424
q3d | 0.5478 | 0.6520
q3e | 0.3641 | 0.8253
q3f | 0.5230 | 0.7131
q3g | 0.8243 | 0.3179
q3h | 0.8601 | 0.2561
q3l | 0.5516 | 0.6339
q3m | 0.6139 | 0.6034
-----------------------------------------------------------
(blanks represent abs(loading)<.3)
Factor rotation matrix
-----------------------------------------
| Factor1 Factor2 Factor3
-------------+---------------------------
Factor1 | 0.8431 0.4610 0.2769
Factor2 | -0.5334 0.7820 0.3224
Factor3 | -0.0680 -0.4195 0.9052
-----------------------------------------
. estat common
Correlation matrix of the varimax rotated common factors
--------------------------------------------
Factors | Factor1 Factor2 Factor3
-------------+------------------------------
Factor1 | 1
Factor2 | 0 1
Factor3 | 0 0 1
--------------------------------------------
. predict FUTURE DYNAMISM DIVERSITY
(regression scoring assumed)
Scoring coefficients (method = regression; based on varimax rotated factors)
--------------------------------------------
Variable | Factor1 Factor2 Factor3
-------------+------------------------------
q3b | 0.07582 0.31198 -0.12150
q3c | 0.00770 0.42358 0.01669
q3d | -0.10003 0.24865 0.00936
q3e | -0.02314 0.12156 0.08707
q3f | 0.13150 -0.02809 0.05570
q3g | 0.36793 0.01214 -0.05684
q3h | 0.47400 -0.04866 0.02340
q3l | -0.00905 0.02469 0.36526
q3m | -0.00374 -0.04271 0.42293
--------------------------------------------
. pwcorr FUTURE DYNAMISM DIVERSITY
| FUTURE DYNAMISM DIVERS~Y
-------------+---------------------------
FUTURE | 1.0000
DYNAMISM | 0.0301 1.0000
DIVERSITY | 0.0418 0.1544 1.0000
.
end of do-file
--
Trevor Zink, MBA, MA
Ph.D. Candidate
UC Regents Special Fellow
Bren School of Environmental Science and Management
University of California, Santa Barbara
[email protected] <mailto:[email protected]>
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/faqs/resources/statalist-faq/
* http://www.ats.ucla.edu/stat/stata/