# Re: st: How to balance an unbalanced panel data set

 From Christian Bustamante To statalist@hsphsun2.harvard.edu Subject Re: st: How to balance an unbalanced panel data set Date Sat, 4 Apr 2009 15:21:15 -0500

```If i'm not wrong, when the panel is balanced, a regression should have
the same number of observations in each group. After doing that you
and Kit suggest me, it still appears the minimum observations per
group equal to 1. Why happens it if with your process the panel is
strongly balanced.

CB

On Sat, Apr 4, 2009 at 2:40 PM, Martin Weiss <martin.weiss1@gmx.de> wrote:
> <>
>
> What is your question exactly?
>
> HTH
> Martin
> _______________________
> ----- Original Message ----- From: "Christian Bustamante" <cdeb77@gmail.com>
> To: <statalist@hsphsun2.harvard.edu>
> Sent: Saturday, April 04, 2009 9:37 PM
> Subject: Re: st: How to balance an unbalanced panel data set
>
>
>> Still appears  Obs per group: min = 1m. Why?
>>
>> On Fri, Apr 3, 2009 at 2:18 AM, Martin Weiss <martin.weiss1@gmx.de> wrote:
>>>
>>> <>
>>>
>>> So the end result is that you end up with only panel 4? If you throw out
>>> observations due to missingness beforehand, that is the inevitable
>>> conclusion...
>>>
>>>
>>> *************
>>> clear*
>>>
>>> inp year id var1
>>> 1998 1 .
>>> 1999 1 .
>>> 2000 1 34
>>> 2001 1 23
>>> 2002 1 25
>>> 1998 2 56
>>> 1999 2 45
>>> 2000 2 67
>>> 2001 2 47
>>> 2002 2 .
>>> 1998 3 11
>>> 1999 3 13
>>> 2000 3 .
>>> 2001 3 .
>>> 2002 3 23
>>> 1998 4 44
>>> 1999 4 24
>>> 2000 4 33
>>> 2001 4 36
>>> 2002 4 41
>>> 1999 5 24
>>> 2000 5 35
>>> 2002 5 45
>>> end
>>>
>>> drop if mi(var1)
>>>
>>> tempvar length
>>> bys id (year): gen `length'=_N
>>> su `length', mean
>>> bys id: drop if `length'!=r(max)
>>> *************
>>>
>>>
>>>
>>> HTH
>>> Martin
>>>
>>> -----Ursprüngliche Nachricht-----
>>> Von: owner-statalist@hsphsun2.harvard.edu
>>> [mailto:owner-statalist@hsphsun2.harvard.edu] Im Auftrag von Christian
>>> Bustamante
>>> Gesendet: Freitag, 3. April 2009 04:15
>>> An: statalist@hsphsun2.harvard.edu
>>> Betreff: Re: st: How to balance an unbalanced panel data set
>>>
>>> Hi Martin
>>> This is a better example:
>>>
>>> ---------------
>>> | year id var1|
>>> |---------------|
>>> | 1998 01 . |
>>> | 1999 01 . |
>>> | 2000 01 34 |
>>> | 2001 01 23 |
>>> | 2002 01 25 |
>>> |---------------|
>>> | 1998 02 56 |
>>> | 1999 02 45 |
>>> | 2000 02 67 |
>>> | 2001 02 47 |
>>> | 2002 02 . |
>>> |---------------|
>>> | 1998 03 11 |
>>> | 1999 03 13 |
>>> | 2000 03 . |
>>> | 2001 03 . |
>>> | 2002 03 23 |
>>> |---------------|
>>> | 1998 04 44 |
>>> | 1999 04 24 |
>>> | 2000 04 33 |
>>> | 2001 04 36 |
>>> | 2002 04 41 |
>>> |---------------|
>>> | 1999 05 24 |
>>> | 2000 05 35 |
>>> | 2002 05 45 |
>>> |---------------|
>>>
>>> In this example I only want the panel with id 04 with data in all the
>>> years, and drop the panels 01, 02 and 03 with a missing value in at
>>> least one year for var1. I also want to drop the panel 05 with doesn't
>>> have values in all years.
>>>
>>> On Thu, Apr 2, 2009 at 4:37 PM, Martin Weiss <martin.weiss1@gmx.de>
>>> wrote:
>>>>
>>>> <>
>>>>
>>>> You ought to give a much clearer excerpt of your data.
>>>>
>>>> If it is really as simple as the one below, use -egen myrowmiss,
>>>
>>> rowmiss()-
>>>>
>>>> and you can easily drop depending on whether the row has any missings by
>>>> -drop if myrowmiss >0 & !mi(myrowmiss)-.
>>>>
>>>>
>>>> HTH
>>>> Martin
>>>> _______________________
>>>> ----- Original Message ----- From: "Christian Bustamante"
>>>
>>> <cdeb77@gmail.com>
>>>>
>>>> To: <statalist@hsphsun2.harvard.edu>
>>>> Sent: Thursday, April 02, 2009 10:50 PM
>>>> Subject: Re: st: How to balance an unbalanced panel data set
>>>>
>>>>
>>>>> Hi and thanks for your response. That's not what I'm looking for. I
>>>>> will be more explicit.
>>>>> Supose the folliwing data set with each row represents a panel
>>>>> observations for a variable:
>>>>>
>>>>> 1 2 3 4 5 6 7 8
>>>>> . . 3 4 5 6 7 8
>>>>> 1 2 3 . 5 6 7 8
>>>>> 1 2 3 4 5 6 7 .
>>>>> 1 2 3 4 5 . . .
>>>>>
>>>>> I want to keep only the observations like the first one and drop the
>>>>> others.
>>>>>
>>>>> Bests
>>>>>
>>>>>
>>>>>
>>>>> On Thu, Apr 2, 2009 at 3:31 PM, Philipp Rehm <philipp.rehm@gmx.de>
>>>>> wrote:
>>>>>>
>>>>>> .
>>>>>>
>>>>>> http://www.stata.com/support/faqs/data/panel.html
>>>>>>
>>>>>> HTH,
>>>>>> Philipp
>>>>>>
>>>>>> Christian Bustamante wrote:
>>>>>>>
>>>>>>> Hi Statalisters,
>>>>>>>
>>>>>>> I have a very unbalanced panel data set, composed for a lot of panel
>>>>>>> and 12 years. I'm interested in "balance" this panel, keeping only
>>>>>>> the
>>>>>>> panel tha have information for a set of variables in all the period.
>>>>>>> How can I do that?
>>>>>>>
>>>>>>> Thanks
>>>>>>>
>>>>>> *
>>>>>> * 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/
>>>>>>
>>>>>
>>>>>
>>>>>
>>>>> --
>>>>> CdeB
>>>>>
>>>>> *
>>>>> * 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/
>>>>
>>>
>>>
>>>
>>> --
>>> CdeB
>>>
>>> *
>>> * 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/
>>>
>>
>>
>>
>> --
>> CdeB
>>
>> *
>> *   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/
>

--
CdeB

*
*   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/
```