Hi,
Following is snapshot of the data

Below is the dataex of what I currently have

Code:
* Example generated by -dataex-. For more info, type help dataex
clear
input str7 poc float date_stata long(tp1 tp2 tp3) float count
"MTU" 22479 50 25  5 1
"MTU" 22479 50 25  5 2
"MTU" 22479 50 25  5 3
"MTU" 22480 60 30 10 1
"MTU" 22480 60 30 10 2
"MTU" 22480 60 30 10 3
"MTU" 22481 70 40 15 1
"MTU" 22481 70 40 15 2
"MTU" 22481 70 40 15 3
end
format %tdMonth_DD,_CCYY date_stata

Now, I want to create a new variable i.e., "new_var" that will have the value of tp1 if count == 1, value of tp2 if count ==2 and so on..

Currently I am showing the data for upto 3 observations for each day, in the real data, the each day has minimum observations of 50 and max observations of 1500, so it is very difficult to write a manual code e.g.,

g new_var = .
replace new_var = tp1 if count == 1
replace new_var = tp2 if count == 2


Any suggestion loop ?

Below is the dataex of what I want. "new_var" is the new variable that I want to generate.

Code:
* Example generated by -dataex-. For more info, type help dataex
clear
input str7 poc float date_stata long(tp1 tp2 tp3) float(count new_var)
"MTU" 22479 50 25  5 1 50
"MTU" 22479 50 25  5 2 25
"MTU" 22479 50 25  5 3  5
"MTU" 22480 60 30 10 1 60
"MTU" 22480 60 30 10 2 30
"MTU" 22480 60 30 10 3 10
"MTU" 22481 70 40 15 1 70
"MTU" 22481 70 40 15 2 40
"MTU" 22481 70 40 15 3 15
end
format %tdMonth_DD,_CCYY date_stata

Thanks