Re: st: loop issue

[Nick Cox <njcoxstata@gmail.com>]

I think it is easier to think of this the other way round. Also, why
think of it as a loop at all? There are just three cases. There is
marginal advantage to setting up a loop over three cases (a loop is at
least three lines long...).

local myvars  var2 var3  var6 var7 var8 var9  var 10

logistic outcome `myvars'
logistic outcome `myvars'  var5
logistic outcome `myvars'  var1 var5

It would be possible to go

local myvars  var2 var3  var6 var7 var8 var9  var 10
local Myvars
foreach v in ""  var5 var1 {
         local Myvars `Myvars' `v'
         logistic outcome `Myvars'
         <something to do with processing results>
}

Constructs such as loops are there to simplify complicated problems,
not to complicate simple problems.

Nick
njcoxstata@gmail.com

On 10 July 2013 10:01, Dherani, Mukesh <M.K.Dherani@liverpool.ac.uk> wrote:

> I have a very basic problem in creating a loop.
> In my data I want to run successive logistic regression models after dropping a few variables. I have developed following code:
>
> local all  var1 var2 var3 ..................... var 10          // include all variables
> local exclvar1 in var2 var3 ................ var10              // exclude var1
> local exclvar15 in var2 var3 ................ var10             // exclude var1 & var5
>         foreach log in all exclvar1 exclvar15 {
>                 eststo `log' {
>
> }
> }
>
> logistic outcome `log'
>
>
> I want stata to produce three different outputs with i) all variables, ii) excluding var1 & iii) excluding var1 & var5. After this it would store the output accordingly. Stata runs the code but only produces one output with "outcome" only.
> Any corrections .... please. [thank you in advance]
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/