[R] NIPALS & categorical variables

[Jim Lemon]

Hi Hana,
If you are using nipals from the plsdepot package, the answer seems to be
maybe. If your categorical variables are at least ordinal level, and you
can arrange the values so that the factor numbering reflects this, it may
be possible to use the numeric values.

"...the variables are assumed to be quantitative, ideally measured on a
more or less continuous scale."

(from http://gastonsanchez.com/plsdepot_nipals.pdf). Obviously you would be
seriously stretching the assumptions of the method, but you will have to
decide whether that is okay.

Jim


On Wed, Dec 16, 2015 at 9:10 PM, Hana Mandová <pmhm at leeds.ac.uk> wrote:

> Hello,
>
> Big help is needed. Can you use nipals for a dataset that contains both
> categorical as well as numerical variables?
> I used nipals instead of pca as some of my numerical variables contain
> missing values, but my categorical variables are complete.
>
> Many many thanks for your help, I've been stuck on this for ages.
>
> Hana
>
> Hana Mandová
> PhD Researcher
> EPSRC Centre for Doctoral Training in Bioenergy
> Faculty of Engineering
> University of Leeds
> Leeds LS2 9JT
> United Kingdom
> E: pmhm at leeds.ac.uk<mailto:pmhm at leeds.ac.uk>
> W: www.engineering.leeds.ac.uk/bioenergy/<
> http://www.engineering.leeds.ac.uk/bioenergy/>
> P Help save paper - every piece counts! Do you really need to print this
> email?
>
>
>         [[alternative HTML version deleted]]
>
>
> ______________________________________________
> R-help at r-project.org mailing list -- To UNSUBSCRIBE and more, see
> 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.
>

	[[alternative HTML version deleted]]