Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: Re: Exclude non-positive values when using collapse


From   Chris Yang <[email protected]>
To   [email protected]
Subject   Re: st: Re: Exclude non-positive values when using collapse
Date   Tue, 21 Jan 2014 22:57:07 -0500

Hi Joseph,

Thanks a lot for your prompt reply.

In my example, I collapse by group, so that for group 1 there are only
four values for var3: 1, 1, -1, 1. If we exclude the -1, the mean
after collapse is (1 + 1 + 1) / 3.

If I understand you correctly, after replacing -1 values with missing
values, the -collapse- command will simply ignore the missing ones
when averaging for variables. Is that correct?

Best regards,

Chris



On Tue, Jan 21, 2014 at 10:44 PM, Joseph Coveney <[email protected]> wrote:
> Chris Yang wrote:
>
> I want to collapse a dataset using the means of the variables.
> However, in my data, some variables may contain value -1, and I want
> to exclude only such values from the mean calculation for the variable
> when collapsing.
>
> For example, given the following dataset:
>
> var1 var2 var3 group
> 0     1      1     1
> 0     -1     1     1
> 1     0      -1    1
> 1     1      1     1
> 1     1      1     2
> 1     0      0     2
>
> After the collapse (by group), for group 1: I want var2 to be (1 + 0 +
> 1) / 3 and var3 to be (1 + 1 + 1) / 3, i.e. ignoring -1 values when
> averaging.
>
> How should I do that? Thanks a lot!
>
> --------------------------------------------------------------------------------
>
> Wait a minute.  What do you want to do here?  Do you want to exclude -1 values
> and calculate the mean of the remaining values, or exclude both zero and -1
> values and calculate the mean of the remaining?  Your first paragraph says the
> first, but your penultimate paragraph says the second (well, sort of--I assume
> there's a typo or two for your var3 expression).  Either way, what I wrote in
> response will work; you just have to revise the missing-value list for
> -mvdecode- and the -if- criterion for the manual loop.
>
> Joseph Coveney
>
> *
> *   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/
*
*   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/


© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index