[R] sem: S is numerically singular: expect problems

John Fox jfox at mcmaster.ca
Tue Apr 9 15:24:30 CEST 2013


Dear Amarnath Bose,

S is the observed-variables covariance or moment matrix and thus doesn't
really have to do with the model that you're fitting, nor with the sem
software per se; rather, S is computed directly from the data. 

It's not possible without the data to know why S is numerically singular.
The message you quote suggests three reasons, but these aren't exhaustive.
Investigating the eigenstructure or SVD of S or the data matrix might help
-- e.g., through a principal components analysis.

I hope this helps,
 John

-----------------------------------------------
John Fox
Senator McMaster Professor of Social Statistics
Department of Sociology
McMaster University
Hamilton, Ontario, Canada


> -----Original Message-----
> From: r-help-bounces at r-project.org [mailto:r-help-bounces at r-
> project.org] On Behalf Of Amarnath Bose
> Sent: Tuesday, April 09, 2013 8:22 AM
> To: r-help at r-project.org
> Subject: [R] sem: S is numerically singular: expect problems
> 
> Dear Users,
> 
> I am a new user of the sem package.
> I have a model that is being flagged by sem as "S is numerically
> singular:
> expect problems"
> 
> I have checked John Fox's response to a similar problem. Obviously the
> variance-covariance matrix is singular, but none of the possible
> reasons
> seems to hold in my case.
> 
> Any leads how I could get the model to work?
> 
> from Prof. John Fox
> 
> That seems to me a reasonably informative error message: The
> observed-variable covariance matrix is singular. This could happen,
> e.g., if
> two observed variables are perfectly correlated, if an observed
> variable had
> 0 variance, or if there were more observed variables than observations.
> 
> Thanks
> 
> 
> --
> 
> 
>      *Amarnath Bose*
> * **Associate Professor                       *
> *Decision Sciences Department*
> *Birla Institute of Management Technology
> *
> Tel:  +91 120 2323001 - 10 Ext.: 398
> Cell: +91 9873179813
> 
> 	[[alternative HTML version deleted]]
> 
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-
> guide.html
> and provide commented, minimal, self-contained, reproducible code.



More information about the R-help mailing list