How to convert categorical variables (name of industries) into letters?

[Lucas Bordure]

Good morning to the community.

I would like to convert automatically categorical variables (names of industries) into letters.

Consumer, Cyclical > CC
Financial > F
Communications > C
Energy > E
Industrial > I
Communications > Co
Consumer, Non-Cyclical > CnC
Utilities > U
Technology > T

* Example generated by -dataex-. To install: ssc install dataex
clear
input str22 industry str3 var2
"Consumer, Cyclical"     "CC" 
"Financial"              "F"  
"Communications"         "C"  
"Energy"                 "E"  
"Industrial"             "I"  
"Communications"         "Co" 
"Industrial"             "I"  
"Industrial"             "I"  
"Consumer, Cyclical"     "CC" 
"Consumer, Cyclical"     "CC" 
"Consumer, Non-cyclical" "CnC"
"Utilities"              ""   
"Financial"              ""   
"Consumer, Cyclical"     ""   
"Financial"              ""   
"Technology"             ""   
"Communications"         ""   
"Consumer, Cyclical"     ""   
"Financial"              ""   
"Financial"              ""   
"Technology"             ""   
"Consumer, Non-cyclical" ""   
"Industrial"             ""   
"Financial"              ""   
"Financial"              ""   
"Industrial"             ""   
"Basic Materials"        ""   
"Communications"         ""   
"Industrial"             ""   
"Industrial"             ""   
"Consumer, Non-cyclical" ""   
"Utilities"              ""   
"Financial"              ""   
"Consumer, Cyclical"     ""   
"Consumer, Non-cyclical" ""   
"Industrial"             ""   
"Communications"         ""   
"Industrial"             ""   
"Basic Materials"        ""   
"Communications"         ""   
"Technology"             ""   
"Industrial"             ""   
"Financial"              ""   
"Financial"              ""   
"Consumer, Non-cyclical" ""   
"Technology"             ""   
"Consumer, Non-cyclical" ""   
"Consumer, Cyclical"     ""   
"Communications"         ""   
"Communications"         ""   
"Communications"         ""   
"Financial"              ""   
"Consumer, Non-cyclical" ""   
"Financial"              ""   
"Technology"             ""   
"Industrial"             ""   
"Industrial"             ""   
"Consumer, Non-cyclical" ""   
"Financial"              ""   
"Consumer, Cyclical"     ""   
"Communications"         ""   
"Basic Materials"        ""   
"Technology"             ""   
"Industrial"             ""   
"Industrial"             ""   
"Consumer, Non-cyclical" ""   
"Industrial"             ""   
"Consumer, Non-cyclical" ""   
"Consumer, Non-cyclical" ""   
"Financial"              ""   
"Consumer, Cyclical"     ""   
"Industrial"             ""   
"Communications"         ""   
"Technology"             ""   
"Utilities"              ""   
"Utilities"              ""   
"Consumer, Cyclical"     ""   
"Financial"              ""   
"Technology"             ""   
"Consumer, Cyclical"     ""   
"Communications"         ""   
"Industrial"             ""   
"Communications"         ""   
"Energy"                 ""   
"Utilities"              ""   
"Financial"              ""   
"Technology"             ""   
"Financial"              ""   
"Energy"                 ""   
"Financial"              ""   
"Consumer, Cyclical"     ""   
"Industrial"             ""   
"Consumer, Cyclical"     ""   
"Consumer, Cyclical"     ""   
"Consumer, Cyclical"     ""   
"Communications"         ""   
"Industrial"             ""   
"Industrial"             ""   
"Basic Materials"        ""   
"Utilities"              ""   
end

Thank you a lot.

[Nick Cox]

gen wanted = cond(strpos(industry, ", Cyc"), "CC", cond(strpos(industry, ", Non"), "CnC", cond(strpos(industry, "Comm"), "Co", substr(industry, 1, 1))))

Writing out several replace statements really isn't wrong!

[Lucas Bordure]

Thank you a lot Nick Cox. The code is working. And it's great to know this formula!