Re: st: is this a missing value issue? (-if- command used)

[Andrea Bennett <mac.stata@gmail.com>]

Many thanks for the inputs!

So, how can I drop a variable such as I have described (containing  
lots of missing values but otherwise values from 0-1) when the value  
of this variable is below a certain threshold, e.g. 0.1? As asked  
before (my previous thread) I cannot use -drop X if Y==Z- because I  
don't want to drop observations but a variable when some condition is  
met.

Many thanks if someone got this done somehow and can share the  
knowledge,

Andrea


On Jul 21, 2008, at 3:22 PM, Johannes Geyer wrote:


> Hey Andrea,
>
> I guess this is a FAQ:
> http://www.stata.com/support/faqs/lang/ifqualifier.html
>
> "if" works like "while" in your example, i.e. in case you have - 
> `var'==.-
> and "missing" is
> a very large number the if-statement does not evaluate to "true" and  
> does
> not execute -drop `var'-. It
> starts evaluating your data with the first observation. If the  
> statement
> is "false", Stata proceeds executing code after the if-statement.
>
> Hope this helps,
>
> Johannes
>
> wner-statalist@hsphsun2.harvard.edu schrieb am 21/07/2008 15:05:02:
>
>
> > Hi!
> >
> > I have the following code in my do-file:
> >
> > qui lookfor zt
> > qui foreach var of varlist `r(varlist)' {
> >   if `var'<=.1  {
> >      drop `var'
> >   }
> > }
> >
> > All my zt* variables are percentages of missing values of the  
> > original
> > variables, e.g. zt_age_2000==0.04 means that 0.04 percent of the
> > variable age in year 2000 are missing. The way I generated the zt-
> > variables, they contain Stata missing values for all years except the
> > corresponding year, e.g. in the above case the zt_age_2000 variable
> > contains missing values for all years smaller and greater than 2000
> > while for the year 2000 it shows the percentage missing.
> >
> > Now with the above query I want to get rid of zt-variable if their
> > missing percentage is below the threshold of 0.1. Strangely, when I
> > use the above commands, it does not perform what I want. However,  
> > when
> > I first recode each zt-variable and set the missing values to zero,
> > then the if-query works fine.
> >
> > So, what am I doing wrong when using the if-command?
> >
> > Many thanks for your considerations,
> >
> > Andrea
> > *
> > *   For searches and help try:
> > *   http://www.stata.com/help.cgi?search
> > *   http://www.stata.com/support/statalist/faq
> > *   http://www.ats.ucla.edu/stat/stata/
> *
> *   For searches and help try:
> *   http://www.stata.com/help.cgi?search
> *   http://www.stata.com/support/statalist/faq
> *   http://www.ats.ucla.edu/stat/stata/
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/