something

[Josh Lamb]

I am an R user trying (and failing) to do something simple in STATA.

I have long form toy data set (see below) where I want to add a numeric variable for each unique combination of `row` and `OPCSDATE_`. However, I would like the numeric variable to start from 1 for unique row value.

* Example generated by -dataex-. For more info, type help dataex
clear
input float row str2 n str9 OPCS_ int OPCSDATE_
 1 "01" "fix1"      23376
 1 "02" "left"      23376
 1 "03" "revise"    23376
 1 "04" "right"     23376
 2 "01" "fix1"      23385
 2 "02" "right"     23385
 3 "01" "cut"       23376
 3 "02" "catheter"  23376
 4 "01" "rev1"      23378
 4 "02" "fix1"      23378
 4 "03" "left"      23378
 4 "04" "fix"       23379
 4 "05" "right"     23379
 5 "01" "abx1"      23379
 6 "01" "decomp1"   23380
 6 "02" "left"      23380
 7 "01" "abx1"      23381
 8 "01" "fix1"      23382
 8 "02" "rev1"      23382
 8 "03" "right"     23383
10 "01" "med1"      23384
10 "02" "hemi"      23384
10 "03" "cut"       23379
10 "04" "bilateral" 23379
11 "01" "cast1"     23385
11 "02" "left"      23385
end
format %tdnn/dd/CCYY OPCSDATE_

I have used

egen float OPCS_dateGrp = group(r OPCSDATE_)

to generate a new variable, but the numeric count does not reset to 1 row each sequential row value.

This is probably quite basic but I have had no luck searching for a solution and would appreciate any help.

[Josh Lamb]

Solution found: Workaround for Lack of Egen Group, by() - Statalist