problem with running MFA
CLaudia Romer-Putz
Hello,
I want to use MFA is to better define the structure of the interrelationships (correlations)
among the variables. I would then use
the factors to run a Cluster Analysis on the score for each forest management
unit.
So this is the structure of my data. Overall, 170 FMUs (or logging concession: I just chopped it). Data frame called IND.
It does not have the Group identification obviously but I added on top as illustration.
I would like to distinguish 6 groups in the analysis, of which #6 will always be supplementary.
|
|
G1 |
G2 |
G3 |
G4 |
G5 |
||||||||
|
FMU |
Area |
Log |
Vol |
Perm |
Regi |
Parent |
Orig |
Int |
Outle |
Subcont |
Mgmt |
Age |
FSC
|
|
1 |
47,700 |
818 |
32,788 |
MoF45 |
Coop |
Amindo |
Ind |
NO |
Dom |
YES |
SVLK |
New |
Y |
|
2 |
24,610 |
590 |
25,000 |
MoF45 |
Coop |
Daya Sakti |
Kor |
NO |
Dom |
YES |
NO |
New |
N |
|
3 |
94,380 |
2,600 |
84,000 |
MoF45 |
Priv |
Kayuma |
Ind |
YES |
Dom |
YES |
NO |
Old |
N |
|
4 |
77,700 |
1500 |
42,610 |
MoF45 |
Priv |
Cental |
Chi |
NO |
Dom |
NO |
NO |
New |
N |
QUESTIONS
1. I am having trouble defining the groups.
2. I want to use the FSC (treated and not) as a supplementary variable.
My code:
MFA(IND, group=c(3,2,3,4,1), type=c("s", rep("n",4)),
ncp=5, name.group=c("Logint","Leg”, “Comp”, “Mark", "FSC"),
num.group.sup=c(5))
> MFA(MODoneRAWNEW, group=c(3,2,3,4,1), type=c("s", rep("n",4)),
+ ncp=3, name.group=c("Logint",”Leg”,”Comp”, "Mark","FSC"),
+ num.group.sup=c(5))
Error in MFA(MODoneRAWNEW, group = c(3,2,3,4,1), type = c("s", rep("n", :
not convenient group definition
THANKS for the help! Claudia Romero
François Husson
MODoneRAWNEW=read.table("MODoneRAWNEW.csv",header=TRUE,sep=",",row.names=1)
res = MFA(MODoneRAWNEW, group=c(2,4,1), type=c("c","n","n"), ncp=5, name.group=c("Logint","Mark","FSC"),num.group.sup=c(3))
Best
FH