I think the desired command is the mean of nonmissing--see -help egen- or try
clear
input id stove refri house
1 1 1 0
2 0 . 1
3 . . 0
end
egen ib=rowmean(stove refri house)
li, noo
On Fri, May 23, 2008 at 12:12 PM, Nick Cox <[email protected]> wrote:
> First principles solution:
>
> gen IB = 0
> foreach v of var stove refri house {
> replace IB = IB + !missing(`v')
> }
>
> Quick and dirty:
>
> egen IB = rownonmiss(stove refri house)
>
> Nick
> [email protected]
>
> Leonor Saravia
>
> I'm learning how to make some programs in Stata and I�m having
> problems in how I could tell Stata to do an indicator I need.
>
> I'm working with the data base for different countries of
> latinamerica. In almost all of them I have the same dummy variables
> which tell me if the interviwed household has or not a particular good
> in the house, but there are cases where the country didn't ask the
> household about having this goods, so in this cases I have the
> variable with missing values.
>
> The dummy variables, which distinguish the goods being studied, are:
> 'stove' (1 if has a stove; 0 other way) ; 'refri' (1 if it has a
> refrigerator ; 0 other way) ; 'house' (1 if it's the owner of the
> house ; 0 other way) .
>
> My problem is that I'd like to perform the next indicator:
>
> gen IB = (stove + refri + house)/q
>
> where 'q' is a number between 1 and 3, depending how many variables
> had values different from missing.
>
> The data looks like this (and I agregate the value - in fraction - for
> the IB that I would like to create):
>
> id_household stove refri house IB
> 1 1 1 0 2/3
> 2 0 . 1 1/2
> 3 . . 0 0/1
>
> I'd appreciate very much if you could help me, because I 'don't know
> how I could tell Stata to do this.
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
