[R] MDS, line of best fit, and id of variables

[Kirsten Green]

David,

I have run the str() function and all my data is int (integer). So I am
trying to change it to numeric using:
str(mortdata)
class(mortdata)
is.numeric(mortdata)
mortdata.num <- as.numeric(data.frame(mortdata))

But I keep getting an error: > mortdata.num <- as.numeric(data.frame(
mortdata))
Error: (list) object cannot be coerced to type 'double

On Tue, Oct 18, 2016 at 1:23 PM, David L Carlson <dcarlson at tamu.edu> wrote:

> What do you get with str(mortdata)? The error message indicates that at
> least one of the variables is not numeric and the second suggests it is a
> factor. You said the values were coded binomially, but they must be
> numeric, e.g. 0, 1 and not "Present" "Absent" or something like that. If
> they are all factors, something like
>
> mortdata1 <- sapply(mortdata, as.numeric)-1
>
> would convert factor levels of 1 and 2 to 0 and 1.
> -------------------------------------
> David L Carlson
> Department of Anthropology
> Texas A&M University
> College Station, TX 77840-4352
>
> -----Original Message-----
> From: R-help [mailto:r-help-bounces at r-project.org] On Behalf Of Kirsten
> Green
> Sent: Tuesday, October 18, 2016 12:28 PM
> To: r-help at r-project.org
> Subject: [R] MDS, line of best fit, and id of variables
>
> Hi,
>
> I have created a dataset that includes 28 rows (burials) and 27 columns
> (variables) that are coded binomially. I have gotten metaMDS to run in the
> pst but now can't seem to get it run at all.
> Error message:
> > mortdata.mds <- metaMDS(mortdata)
> Error in FUN(X[[i]], ...) :
>   only defined on a data frame with all numeric variables
> In addition: Warning message:
> In Ops.factor(left, right) : ‘<’ not meaningful for factors
>
> I'd like to create a 3D MDS plot and add the line of best fit for the 3
> dimensions (3 variables). I am also trying to figure out, or understand,
> which variables are causing the variation.
>
> I ran PCA and it told me that with 3 variables approximately 50% of the
> data variation is explained. So I assumed that meant that running MDS in 3
> dimensions would show me 3 variables causing the variation but I can't get
> that to work.
>
> Here is my code so far (i've also attached it to the email):
>
> mortdata<-read.csv("Table5.5.csv", header=TRUE)
> mortdata
> row.names(mortdata) <- mortdata[,1]
> mortdata <- mortdata[,-1]
> mortdata
>
> mortdata.mds <- metaMDS(mortdata)
> mortdata.mds.alt <- metaMDS(mortdata, distance="euclidean", k=3, trymax=50,
>       autotransform=FALSE, noshare=FALSE)
>
>
> *object = mortdata.mds.alt
>
> names(mortdata.mds.alt)
>
> mortdata.mds.alt
> summary(mortdata.mds.alt)
>
> *stress plot
> stressplot(mortdata.mds.alt)
>
> x <- mortdata.mds.alt$species
> y <- mortdata.mds.alt$points
> na.exclude(mortdata.mds.alt)
> vScoresScale <- scale(, center = TRUE, scale = TRUE)
>
>
> plot(mortdata.mds.alt)
> plot(mortdata.mds.alt, type="t")
>
> *multiple linear regression model
> lm(formula = x ~ y)
> abline(lm(x ~ y), col="red")
>
>
> *scatterplot3D
> library(scatterplot3d)
> attach(mortdata.mds.alt)
> scatterplot3d(mortdata.mds.alt, x="sampleScores", y="variableScores",
> main="3D Scatterplot")
>
> Any help would be greatly appreciated. I can also send the dataset if that
> helps.
>
> --
> *Kirsten Green*
> kagbones at gmail.com
> 916-712-5193
> ______________________________________________
> 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.
>



-- 
*Kirsten Green*
kagbones at gmail.com
916-712-5193

	[[alternative HTML version deleted]]