extracting column number based on minimum row value

[Adhikari Pramod]

Hello
I have a sample data set with four columns (L1 to L4) and five rows. I am trying to extract the column number which contains the minimum value across the rows. If there are two rows with the same minimum value, I would like to extract the first column that contains the minimum value (ignoring the other columns).

. version
version 13.1

. clist L1-L4

            L1         L2         L3         L4
  1.         2          2          .          .
  2.         3          2          .          2
  3.         7          3          6          8
  4.         4          3          7          3
  5.         4          5          7          9

I then calculate the row minimum.

. egen min=rowmin(L1 L2  L3  L4)

.clist

            L1         L2         L3         L4        min
  1.         2          2          .          .          2
  2.         3          2          .          2          2
  3.         7          3          6          8          3
  4.         4          3          7          3          3
  5.         4          5          7          9          4

Now I would like to extract the column number which contains the minimum value. If there are two minimum values in a row, the column containing the first minimum value is what I am after,
I tried the following:

 

. gen col_no=.
(5 missing values generated)

. local array1 "L1 L2 L3 L4" 

. local i=1

. local n : word count `array1'

. while `i' <= `n' {
  2.          local var1 : word `i' of `array1'
  3.                  replace col_no =`i' if `var1'==min
  4.                  local i = `i' + 1
  5.         }               
(2 real changes made)
(4 real changes made)
(0 real changes made)
(2 real changes made)

. clist

            L1         L2         L3         L4        min     col_no
  1.         2          2          .          .          2          2
  2.         3          2          .          2          2          4
  3.         7          3          6          8          3          2
  4.         4          3          7          3          3          4
  5.         4          5          7          9          4          1

The output is not exactly what I wanted. For rows 1, 2, and 4, the extracted column numbers should be 1, 2 and 2 respectively.

Thanks in advance.

[Joro Kolev]

I cannot import your data and try it, but something like this should do the trick:

gen col_no = 1 if L1==min
replace col_no = 2 if L2==min & missing(col_no)
replace col_no= 3 if L3==min & missing(col_no)
replace col_no=4 if L4==min & missing(col_no)

[Adhikari Pramod]

Thanks Joro. This solves the problem I have.