Re: st: Adding a variable to a panel dataset

[Joerg Luedicke <joerg.luedicke@gmail.com>]

As I said before, you need to -reshape- your data such that you have a
dataset with 77 (7 countries x 11 years each) rows and 3 variables
(your country ID, year, and unemployment rate. Then you can merge by
country ID and year. Have a look at -help reshape- and -help merge-
and study the examples therein. Perhaps also have a look at the manual
entries for these commands.

J.

On Tue, Oct 2, 2012 at 12:11 PM, Jason Hecht <jayhstata@gmail.com> wrote:
> Joerg, Chris -
>
> I initially created the panel out of individual years.  In doing so, I
> did, in fact, create a new variable called "nation" that exists for
> each observation in the full dataset (e.g., "China" is repeated for
> each firm year observation for any firm located in China).
>
> So now I just need a way to get the unemployment rate for each country
> (from 1998 to 2008) slotted into the appropriate firm-level panel.
>
> Can I use my "nation" variable to do this?  I believe I should be able
> to merge the 11-line dataset (with the country unemployment rates)
> into the full dataset.  For example, I can use "China" (a name in the
> 11-line 'text' dataset) to match with "China" which also appears in
> the "nation" variable in the full dataset.
>
> Is this the way to proceed?
>
> Thanks so very much for your help.
>
> Jason
>
>
> On Tue, Oct 2, 2012 at 10:12 AM, Joerg Luedicke
> <joerg.luedicke@gmail.com> wrote:
>> If I understand the problem correctly, you will need to -reshape long-
>> (see -help reshape-) your country dataset. You will then need a
>> country identifier that is the same in both of your datasets. If you
>> don't have such an ID yet you may want to create one. Then you can
>> simply merge country and firm-level data by this ID and year.
>>
>> J.
>>
>> On Mon, Oct 1, 2012 at 11:39 PM, Jason Hecht <jayhstata@gmail.com> wrote:
>>> Chris -
>>>
>>> A point of clarification:
>>>
>>> In the example that you provided, it appears that the "wpid" file has
>>> as many obs. as the "employee" file.
>>>
>>> My problem is that I have 7 different country dummy variables that
>>> identify where the individual firm is located.  However, I want to
>>> create a single variable that will have the unemployment rate (from
>>> 1998 to 2008) mapped into each firm's panel.
>>>
>>> So let's say my "wpid" is called "unrate" (a text file) which will
>>> have 11 lines and 14 columns arrayed as:
>>>
>>> 1998  China 8.9  Fra  6.4 . . . US  5.4
>>> 1999  China 8.3  Fra  5.8 . . . US  4.5
>>> .              .              .      .            .     .
>>> .              .              .      .            .     .
>>> .              .              .      .            .     .
>>> 2008 China 7.2   Fra  8.2. . .  US   7.5
>>>
>>> where the above variables would be named as "year" "China" "RUNCChn"
>>> "Fra" RUNCFra" .... "US" "RUNCUS"
>>>
>>> Now I would need to merge the above dataset with my analagous
>>> "employee" dataset which has 155,386 firm-year panel observations
>>> (firm id #'s go from 1 to 14,126 and year (1998-2008) is a variable.
>>>
>>> I'm not sure if your FAQ example would work because I need to merge
>>> unempoyment rates for seven different countries into the firm-level
>>> panels.
>>>
>>> I'm sorry if I am misinterpreting the FAQ example.  I'm relatively new
>>> to STATA and am learning how to use the programming commands.
>>>
>>> Kindly advise and thanks so very much for your help.
>>>
>>> Jason
>>>
>>> On Mon, Oct 1, 2012 at 7:04 AM, Christopher Baum <kit.baum@bc.edu> wrote:
>>>> On Oct 1, 2012, at 2:33 AM, Jason wrote:
>>>>
>>>>> I have constructed a panel dataset containing data on +14,000 firms
>>>>> from 1998 to 2008.  I created a dummy variable that identifies the
>>>>> country where each firm is located (i.e., China, France, Germany,
>>>>> Japan, India, UK, or US).
>>>>>
>>>>> I now want to add the unemployment rate for each country that would
>>>>> map into each firm's panel. That is, I would like a single column that
>>>>> would match the country's unemployment rate to the panel of the
>>>>> individual firm.
>>>>>
>>>>> Is there a relatively easy way to do this?
>>>>
>>>> This is a FAQ:
>>>>
>>>> http://www.stata.com/support/faqs/data-management/group-characteristics-for-subsets/
>>>>
>>>> Kit
>>>>
>>>>
>>>> Kit Baum   |   Boston College Economics & DIW Berlin   |   http://ideas.repec.org/e/pba1.html
>>>>                              An Introduction to Stata Programming  |   http://www.stata-press.com/books/isp.html
>>>>   An Introduction to Modern Econometrics Using Stata  |   http://www.stata-press.com/books/imeus.html
>>>>
>>>>
>>>> *
>>>> *   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/
>> *
>> *   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/
*
*   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/