Help needed

[Mari-Ali Pri]

Hi, thank you in advanced for helping me, I'm doing an internship and i'm really struggling with STATA cause i've never worked with it. So i've cleaned my data (I had 3 datasets and i've append them) but now I want to create groups for triplets (so there is a case and two controls) and I want to analyse within the groups of triplets. They all have the same number but not the same letters so im guessing with strings but i really don't know. Example : EX-E1 , EX-HHC1 et EX-VC1 are together. Can someone help me? Thank you so much

[Andrew Musau]

They all have the same number but not the same letters

Do the numbers always appear at the end of the string? If not, present a data example using the dataex command (see FAQ Advice #12 for details).

clear
input str10 ID
"EX-E1"  
"RK-32"
"EX-HHC1" 
"EX-VC1"
"ML13"
"V6"
"TPP13"
"MX32"
end

gen group= ustrregexra(ID, "(.*)[^\d]([\d]+$)", "$2")

Res.:

. sort group

. l, sepby(group)

     +-----------------+
     |      ID   group |
     |-----------------|
  1. |  EX-VC1       1 |
  2. |   EX-E1       1 |
  3. | EX-HHC1       1 |
     |-----------------|
  4. |   TPP13      13 |
  5. |    ML13      13 |
     |-----------------|
  6. |   RK-32      32 |
  7. |    MX32      32 |
     |-----------------|
  8. |      V6       6 |
     +-----------------+

[Mari-Ali Pri]

Hi, yes the numbers are always at the end. So i'm not sure if I can use the code you sent. This is a real example of my data.
AUDEM-E1
AUDEM-VC1
AUDEM-HHC1
AUDEMHHC-2
AUDEM-VC2
AUDEM-E2
ETC, ETC
I want all the 1 together so it makes a triplet. Thanks!

[Dirk Enzmann]

Mari-Ali Pri : Welcome to the Stata Forum!

You have been asked to read the FAQ Advice #12 and to present a data example using the dataex command -- please do so (I would even recommend to read the complete FAQ = advice on posting to Statalist). There you will also see that (for good reasons) you are advised not to post attachments (see FAQ Advice #12.5).