Skip to content

PCA analysis and bootstraped loadings

4 messages · stephen sefick, William Revelle, Efstathia Defteraiou

#
Dear All,

I am relatively new in R.
Im working with the 'psych' package and 'principal' function.
I would like to know how to generate the bootstraped conf.intervals  
for loadings,
looking for sth similar to setting 'n.iter' argument for the 'fa' function.

If in 'psych' can't work and suggest me the 'boot' package please  
provide specific Rscript since I don't understand the commands and  
arguments that have to be used before calling the function 'boot'(  
what are indices? what to define as what inside function(){})

The names Im using are included in the following code:
'newdata3.1' is my data and provided as data.frame

makingtheanalysis3.1 <-principal(newdata3.1, nfactors =3,
                               residuals = FALSE,
                               covar=FALSE,rotate="varimax",scores=TRUE)


I am sorry for not providing a specific code but my data are too large

Any Help appreciated
Cheers!
#
Hi,

Please search the mailing list archives for this, or type bootstrapped PCA
R into google. Please provide a minimal self-contained example of what you
are trying to solve. Please read the posting guide that is referenced at
the end of every email.
kind regards,

Stephen

On Mon, Apr 13, 2015 at 11:07 AM, Efstathia Defteraiou <
Efstathia.Defteraiou at student.uibk.ac.at> wrote:

            

  
    
#
psych does not currently have bootstrapped confidence intervals for loadings.  That is a reasonable request and I will try to add it, perhaps in the ?real soon now? version of 1.5.4 (almost finished), perhaps in the next release,

Bill
William Revelle		           http://personality-project.org/revelle.html
Professor			           http://personality-project.org
Department of Psychology   http://www.wcas.northwestern.edu/psych/
Northwestern University	   http://www.northwestern.edu/
Use R for psychology             http://personality-project.org/r
It is 3 minutes to midnight	   http://www.thebulletin.org
#
Dear All,
Thank You for the quick responses.
Managed to solve my problem through:
http://www.faculty.biol.ttu.edu/strauss/multivar/R/SamplePCABootstrap.R.txt
or
http://r.789695.n4.nabble.com/bootstrapped-eigenvector-method-following-prcomp-td877655.html
Used the first one however, code is too long since everything is  
manually done.
The suggestion of Bill would be very kind and save a lot of time in  
the future.
Thanks William for clearing this up.

Cheers
Efi


Zitat von William Revelle <lists at revelle.net>: