Hi All,
I need to fit a mutlivariate normal model to a dataset in order to obtain the mean and covariance parameters. I see that the MVN function in the MCLUST package can do this, however when I try to run even the simplest example provided in the documentation, as below, I get the following error:
n <- 1000
set.seed(0)
x <- rnorm(n, mean = -1, sd = 2)
mvn(modelName = "X", x)
Error in .Fortran("mvn1d", as.double(data), as.integer(n), double(1), :
Fortran symbol name "mvn1d" not in DLL for package "mclust"
I cannot see a resolution to this in the list, and wonder whether I need to load something else separately to the package? Could it be that I need to load a separate DLL?
Many thanks for your help.
Natalie
(PhD Student in Medical Statistics at Reading University)
Error using function MVN in package MCLUST: Fortran symbol name not in DLL for package
3 messages · Natalie Franklin, Sarah Goslee, Brian Ripley
I suspect at a minimum we need the basic information requested in the posting guide: your OS, version of R< version of the package in question. sessionInfo() is a useful command. You shouldn't need to load anything beyond the package itself. Sarah On Fri, Dec 9, 2011 at 4:01 AM, Natalie Franklin
<N.J.Franklin at pgr.reading.ac.uk> wrote:
Hi All,
I need to fit a mutlivariate normal model to a dataset in order to obtain the mean and covariance parameters. I see that the MVN function in the MCLUST package can do this, however when I try to run even the simplest example provided in the documentation, as below, I get the following error:
n <- 1000
set.seed(0)
x <- rnorm(n, mean = -1, sd = 2)
mvn(modelName = "X", x)
Error in .Fortran("mvn1d", as.double(data), as.integer(n), double(1), ?:
?Fortran symbol name "mvn1d" not in DLL for package "mclust"
I cannot see a resolution to this in the list, and wonder whether I need to load something else separately to the package? Could it be that I need to load a separate DLL?
Many thanks for your help.
Natalie
Sarah Goslee http://www.functionaldiversity.org
On Fri, 9 Dec 2011, Natalie Franklin wrote:
Hi All,
I need to fit a mutlivariate normal model to a dataset in order to
obtain the mean and covariance parameters. I see that the MVN
function in the MCLUST package can do this, however when I try to
run even the simplest example provided in the documentation, as
below, I get the following error:
n <- 1000
set.seed(0)
x <- rnorm(n, mean = -1, sd = 2)
mvn(modelName = "X", x)
Error in .Fortran("mvn1d", as.double(data), as.integer(n), double(1), :
Fortran symbol name "mvn1d" not in DLL for package "mclust"
I cannot see a resolution to this in the list, and wonder whether I
need to load something else separately to the package? Could it be
that I need to load a separate DLL?
Please the footer of this message. It works for me, and we need the 'at a minimum' information we asked for to begin to guess what might be wrong for you. However, this is pretty much overkill: there is nothing 'mutlivariate' about your example, and even if there were mean() and var() give the standard parameter estimates.
Many thanks for your help. Natalie (PhD Student in Medical Statistics at Reading University)
______________________________________________ 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.
Brian D. Ripley, ripley at stats.ox.ac.uk Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/ University of Oxford, Tel: +44 1865 272861 (self) 1 South Parks Road, +44 1865 272866 (PA) Oxford OX1 3TG, UK Fax: +44 1865 272595