[R-marketing-bugs] error chap 11. Mclust()

Chris Chapman cnchapman at gmail.com
Mon Aug 24 03:06:51 CEST 2015


It looks like the recent update of Mclust to version 5 has caused this to
fail.

If you are using a Mac or Linux and can compile packages from source, then
you could downgrade Mclust to the previous version 4.4 as follows:

old.mc <- "
https://cran.r-project.org/src/contrib/Archive/mclust/mclust_4.4.tar.gz"

install.packages(old.mc, repos=NULL, type="source")

If you use Windows you  might be able to find a compiled ZIP for mclust 4.4
on the web somewhere.


After that, everything works as expected:

> seg.mc <- Mclust(seg.df.num)
> summary(seg.mc)
----------------------------------------------------
Gaussian finite mixture model fitted by EM algorithm
----------------------------------------------------

Mclust EEV (ellipsoidal, equal volume and shape) model with 3 components:

 log.likelihood   n df       BIC       ICL
      -5256.222 300 71 -10917.41 -10955.48

Clustering table:
  1   2   3
111 115  74


Hopefully the Mclust developers will fix this. I'll email them and ask!
Thanks for the report, and I hope you're enjoying the book,

-- Chris


On Sun, Aug 23, 2015 at 5:38 PM, Caleb Terrel Orellana <
caleb.terrel at gmail.com> wrote:

> Hi , on page 314 of the book R MARKETING  RESEARCH AND ANALYTICS
>
> the code:
>
> library(mclust)
> seg.mc <- Mclust(seg.df.num)
>
> result:
>
> Error in svd(shape.o, nu=0) :infinite or missing values in 'x'.
>
> why?
>
>
> --
>
> Saludos cordiales.
>
> *Caleb Terrel Orellana*
> CONSULTOR ESTADÍSTICO
> 993 319 507
> http://pe.linkedin.com/pub/caleb-terrel-orellana/29/a20/933
> https://www.facebook.com/pages/Statistical-Solutions/1559961577550284
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.r-forge.r-project.org/pipermail/r-marketing-bugs/attachments/20150823/30158d8c/attachment.html>


More information about the R-marketing-bugs mailing list