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: How to drop individuals from sample?
From
Dheepan Ratha Krishnan <[email protected]>
To
[email protected]
Subject
Re: st: How to drop individuals from sample?
Date
Wed, 25 Jul 2012 21:21:54 +1000
-egen- worked well for me. thanks for the advice.
On 7/25/12, Nick Cox <[email protected]> wrote:
> In addition, -egen- is happy with arguments such as -total(event ==
> 1)-, so the code can be simplified by not creating an indicator
> variable.
>
> Nick
>
> On 25 Jul 2012, at 08:25, Maarten Buis <[email protected]> wrote:
>
>> On Wed, Jul 25, 2012 at 8:54 AM, Dheepan Ratha Krishnan wrote:
>>> I am looking for a stata
>>> command that will enable me to exclude individuals who have
>>> experienced a particular life event more than once in my dataset.
>>
>> That depends on how your data is structured. I will assume that the
>> data has the following structure: Each event has its own record, so
>> the same individual can have multiple records. The type of event is
>> stored in a separate variable, for example:
>>
>> person event_date event_type other_vars
>> 1 02Jan2012 married ...
>> 1 03Jan2012 divorced ...
>> 1 04Jan2012 married
>> 2 01Feb1950 married ...
>> 2 01Jun2012 widowed ....
>>
>> In this case you would want to ignore person 1 as he married twice (in
>> three days...). Typically you do not have the variable event_type as a
>> string but as a numeric variable with labels attached. Say 1 is
>> married 2 is divorced and 3 is widowed, and you are interested in the
>> event married. Than you type:
>>
>> gen byte married = event == 1
>> bys person : egen tot_married = total(married)
>> drop married
>>
>> The variable tot_married counts the number of times a person got
>> married, so you want to ignore all persons with tot_married > 2. So
>> you add to any subsequent command the condition -if tot_married < 2-.
>>
>> This way you ignore the entire person 1, but say you want to use the
>> first marriage of person 1 just not any subsequent marriages. Than you
>> type:
>>
>> gen byte married = event == 1
>> bys person (event_date) : gen tot_married = sum(married)
>> drop married
>>
>> To read more about this see:
>> http://www.stata.com/support/faqs/data-management/true-and-false/
>> help egen
>> help sum()
>> help if
> *
> * 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/