0
votes

I am running an analysis with FactominR. For my data, I need to run a FAMD because I have mixed continuous and categorical variables. I have active continuous and categorical variables, and supplementary categorical variables. I can run the FAMD, and plot the results. But I don't find how to plot the supplementary variables on the graphics. The summary shows that the supplementary variables were included in the analysis. I did a mooc on this package and played around a lot with the different functions, and know how to do this for others like PCA, MFA, ... but I don't find how to display them for the FAMD I expected a graph similar to the MFA for the different modalities.

I have read about the plot.FAMD function and it does not seem that the possibility of displaying or not the supplementary variable exist. Which I find strange because for the MFA , plot includes the supplementary variable and one can display them or not.

The example included in the package does not have supplementary variable. And it runs fine but I can't compare.

# This works except the supplementary variable are invisible on all graphs
res <- FAMD(luc, ncp = Inf, sup.var = c(1,7,8,10,11,13,14))
summary(res, nbelements=Inf)

# Here are the option to display the different graphs from FAMD
plot(res, choix="var")
plot(res, choix="quanti")
plot(res, choix="quali")
plot(res, choix = "ind")

Error in match.arg(choix, c("ind", "var", "quanti", "quali")) : 
  'arg' should be one of “ind”, “var”, “quanti”, “quali”

#what I would do/did with an MFA to remove the suppl variables on the ind factor map
plot(res, invisible=c("var","quali.sup"))

#Same on the Variable graph for an MFA
plot(res, choix="var",invisible=c("quali.sup","quanti.sup"))
1

1 Answers

1
votes

I found out the problem. I was using an older version of FactomineR 1.41 update to 1.42 solved it.

If you have the same issue, and want to test, you can used the following code given to me by François Husson one of the developper.

data(wine)
res <- FAMD(wine[,c(1,2,30,31)],sup.var=1)
plot(res)                  ## graphe avec individus + modalités actves + modalités supplémentaires
plot(res,invisible=c("ind","quali")) ## graphe avec modalités supplémentaires (on rend invisible les individus et modalités actives)
plot(res,choix="var")     # graphe avec toutes les variables (sumbole différent pour les variables supplémentaires)