round and trailing zero

I have a "result":

  > hazards

          (60, 70]    (70, 80]    (80, 90]   (90, 100]
[1,] 0.046612937 0.115643783 0.273613266 0.450127975

Two issues: (i) Too many decimals, and (ii) it seems to be an 1x4
matrix, I only need the first row. (i):

  > haz <- round(hazards, 3)
  > haz

       (60, 70] (70, 80] (80, 90] (90, 100]
[1,]    0.047    0.116    0.274      0.45

As expected, the fourth element lost a trailing zero. I'll deal with
that, but first (ii):

  > haz[1, ]

   (60, 70]  (70, 80]  (80, 90] (90, 100]
      0.047     0.116     0.274     0.450

And the trailing zero is mysteriously recovered!

Is there some general rule governing this behaviour?

29 juli 2024 kl. 16:23 skrev Duncan Murdoch <murdoch.duncan at gmail.com>:

?On 2024-07-29 10:06 a.m., G?ran Brostr?m wrote:

I have a "result":

 > hazards

         (60, 70]    (70, 80]    (80, 90]   (90, 100]
[1,] 0.046612937 0.115643783 0.273613266 0.450127975
Two issues: (i) Too many decimals, and (ii) it seems to be an 1x4
matrix, I only need the first row. (i):

 > haz <- round(hazards, 3)
 > haz

      (60, 70] (70, 80] (80, 90] (90, 100]
[1,]    0.047    0.116    0.274      0.45
As expected, the fourth element lost a trailing zero. I'll deal with
that, but first (ii):

 > haz[1, ]

  (60, 70]  (70, 80]  (80, 90] (90, 100]
     0.047     0.116     0.274     0.450
And the trailing zero is mysteriously recovered!
Is there some general rule governing this behaviour?

R uses the same format for every element in each column when printing a matrix or dataframe, and for every element in a vector.

Your first example had only one element per column.  If you had printed t(haz) you'd get numbers displayed like the second version, where haz[1,] converts that row to a vector.

Duncan Murdoch