I am performing a factor analysis on a 524 by 211 matrix. I'm using the bart_spher()
, KMO()
, and eigen(cor())
functions before running factanal()
. My matrix is "doubles" of 1-7 with no NA values.
When I run the three functions on the full matrix, I get the following error:
Error in eigen(cor(Factors)) : infinite or missing values in 'x'
In addition: Warning messages:
1: In cor(Factors) : the standard deviation is zero
2: In cor(x = x, use = use) : the standard deviation is zero
3: In cor(r, use = "pairwise") : the standard deviation is zero
4: In cov2cor(Q) :
diag(.) had 0 or NA entries; non-finite result is doubtful
5: In cor(Factors) : the standard deviation is zero
I also end up getting NA's for the Bartlett and KMO numbers. But there are no infinite or NA values!
If I use only the first 113 columns (a matrix of 524 by 113), then everything works out fine. Is there a column limit I am unaware of for eigen(cor())
?