Ambiguous abbreviation in a loop problem

[Sandeep]

Hello all,

i have this code,

forvalues i=97(2)99{;
generate Wife`i'=.;
generate Head`i'=.;
};

forvalues i=97(2)99{;
replace Wife`i'=1 if (SN`i'==2 & (Rel`i'==20 | Rel`i'==22));
replace Head`i'=1 if SN`i'==1 & Rel`i'==10;
};

My problem is that it creates the Wife97.......Head99 variables fine. But when it reaches the second loop, it says that Wife is a "Ambiguous abbreviation". I can't seem to figure out why. I tried setting the 'varabbrev off' but if I do that, in the second loop it says that the variables do not exist. How can I get around this? Thank you for your help.

[Nick Cox]

I am not clear what is happening but a report on

describe Wife*

may help.

I am clear that you should not need two loops here. Your code collapses to

forvalues i = 97(2)99 {
   gen Wife`i' = 1 if (SN`i'==2 & (Rel`i'==20 | Rel`i'==22))
   gen Head`i' = 1 if SN`i'==1 & Rel`i'==10 
}

where the question of line delimiter is here set on one side. It is a matter of taste, only, but I find this more explicit:

 
foreach i in 97 99 {
   gen Wife`i' = 1 if (SN`i'==2 & (Rel`i'==20 | Rel`i'==22))
   gen Head`i' = 1 if SN`i'==1 & Rel`i'==10 
}

[William Lisowski]

I'll hazard a guess: in the original code, as opposed to what was typed above, there is a space between "Wife" and "`i'" so that Stata is attempting to interpret the command

replace Wife `i'=1 if (SN`i'==2 & (Rel`i'==20 | Rel`i'==22));

When I simulate that, I get the same error message.

But in any event, Nick's elegant solution is to be preferred.