'x' and 'y' lengths differ

Question

I am trying to determine optimal number of clusters.

# Determine optimal number of clusters
wss<-rep(0,2)
wss[1]<-sum(scale(price[,2:2],scale=FALSE)^2)
for(i in 2:16)
wss[i]<-sum(kmeans(price[,2:2],centers=i)$withinss)
plot(4:2,wss,type="b",xlab="Number of clusters",ylab="Within-cluster sum of squares")

Every line works except the last one. The last one gives an error:

Error in xy.coords(x, y, xlabel, ylabel, log) : 'x' and 'y' lengths differ

I have tried some solutions from other questions but no luck. Any advice? Thanks a tons!!

SAMPLE DATA:

    Country        Price

    Albania        1.57
    Andorra        1.24
    Azerbaijan     0.47
    Austria        1.33
    Belarus        0.73
    Belgium        1.54
    Bosnia & Herz. 1.29
    Bulgaria       1.13
    Croatia        1.44
    Czech Rep.     1.32
    Cyprus         1.28
    Denmark        1.74
    Estonia        1.41
    Finland        1.61
    France         1.67
    Georgia        0.9

Welcome to SO. You need to share sample data with dput and the expected output — Sonny
Afghanistan 0,67 Albania 1,57 Algeria 0,35 Andorra 1,24 Angola 0,52 This is sample data, there are actually 47 rows. I would expect it to have a "curve" as an output to see what is the optimal clustering count. — Denis
use dput(yourData) or dput(head(yourData)) to share data — Iman

Lostman Lostman · Accepted Answer · 2019-04-14T10:49:43

Length of wss (y variable) is 16 but in x-axis you are using 4:2 (length is 3). That's why you are getting the Error.

change 4:2 to 17:2 to make length of x and y variables same. Like:

plot(17:2,wss,type="b",xlab="Number of clusters",ylab="Within-cluster sum of squares")

'x' and 'y' lengths differ

1 Answers