R - Svd() Function - Infinite or Missing Values in 'X'

R - svd() function - infinite or missing values in 'x'

You have columns composed of all zeroes. Using scale on a column of all zeroes returns a column composed of NaN. To solve this, remove columns where you have all zeroes (svd will not reveal anything new about them), or replace NaN columns with zero after using scale.

Reproducible example:

mat <- matrix(c(1,2,3,0,0,0,2,4,6,5,12,13),nrow = 3)
     # [,1] [,2] [,3] [,4]
# [1,]    1    0    2    5
# [2,]    2    0    4   12
# [3,]    3    0    6   13
scale(mat)
     # [,1] [,2] [,3]       [,4]
# [1,]   -1  NaN   -1 -1.1470787
# [2,]    0  NaN    0  0.4588315
# [3,]    1  NaN    1  0.6882472
# attr(,"scaled:center")
# [1]  2  0  4 10
# attr(,"scaled:scale")
# [1] 1.000000 0.000000 2.000000 4.358899
svd(mat) #fine
svd(scale(mat)) # not fine

Error in svd(x, nu = 0, nv = k) : infinite or missing values in 'x'. There are no NA or Inf values in matrix

NAs are tricky. Consider this:

> a = c(1, NA, 2)
> which(a == NA)
integer(0)

> a == NA
[1] NA NA NA

Equality checks with NA will result in NA. The proper way of checking for NAs is with the is.na() function:

> is.na(a)
[1] FALSE  TRUE FALSE

There's also is.infinite() for the case of Inf, though in that case direct comparison works (e.g. (1/0) == Inf yields TRUE).

Related Topics