Announcement

Collapse
No announcement yet.
X
  • Filter
  • Time
  • Show
Clear All
new posts

  • How to read the results of "detail" option of "summarize" syntax properly?

    When running the detail option for the summarize syntax for the numeric variable "Sheet2", I have the results below:
    Click image for larger version

Name:	Capture.PNG
Views:	1
Size:	10.8 KB
ID:	1681728


    I am wondering how to read the results. In particular, I stuck in
    1. reading what does "Sum of wgt." mean ?
    2. Why the largest of 95 percentiles (1123875) is larger than the smallest of 99 percentiles(5664733)
    3. How to read the results properly, e.g. whether in the 1% percentile, 905 is the highest value and 274 is the lowest value?

  • #2
    "Smallest" is independent of the other column and lists the 5 smallest values in ascending order. Same with largest.

    "Sum of wgt." only applies if you use the command with weights. If not, its identical to the Obs (each observation has the weight of 1).
    Best wishes

    (Stata 16.1 MP)

    Comment


    • #3
      Originally posted by Felix Bittmann View Post
      "Smallest" is independent of the other column and lists the 5 smallest values in ascending order. Same with largest.

      "Sum of wgt." only applies if you use the command with weights. If not, its identical to the Obs (each observation has the weight of 1).
      Thanks Felix Bittmann , I am wondering what does the number in the Percentiles column mean?
      Best regards.

      Comment


      • #4
        The smallest 4 values (not 5) and largest 4 (ditto) values are shown in the second column.

        As you have 10710 values, the 99th percentile 1123875 is smaller than each of the largest 4 values listed.

        Comment


        • #5
          Originally posted by Nick Cox View Post
          The smallest 4 values (not 5) and largest 4 (ditto) values are shown in the second column.

          As you have 10710 values, the 99th percentile 1123875 is smaller than each of the largest 4 values listed.
          I got it, thanks Nick Cox.

          Comment

          Working...
          X