Your data structure is now clearer. Thanks for that. I gather that you
mean "add".
Here is one way to proceed:
. l
+-----------------------+
| month sales brand |
|-----------------------|
1. | 1 25 1 |
2. | 2 13 1 |
3. | 3 19 1 |
4. | 4 23 1 |
5. | 1 21 2 |
|-----------------------|
6. | 2 17 2 |
7. | 3 19 2 |
8. | 4 26 2 |
9. | 1 11 3 |
10. | 2 19 3 |
|-----------------------|
11. | 3 19 3 |
12. | 4 28 3 |
+-----------------------+
. reshape wide sales, i(month) j(brand)
(note: j = 1 2 3)
Data long -> wide
------------------------------------------------------------------------
-----
Number of obs. 12 -> 4
Number of variables 3 -> 4
j variable (3 values) brand -> (dropped)
xij variables:
sales -> sales1 sales2 sales3
------------------------------------------------------------------------
-----
. l
+----------------------------------+
| month sales1 sales2 sales3 |
|----------------------------------|
1. | 1 25 21 11 |
2. | 2 13 17 19 |
3. | 3 19 19 19 |
4. | 4 23 26 28 |
+----------------------------------+
. gen sales1and2 = sales1 + sales2
. l
+---------------------------------------------+
| month sales1 sales2 sales3 sales1~2 |
|---------------------------------------------|
1. | 1 25 21 11 46 |
2. | 2 13 17 19 30 |
3. | 3 19 19 19 38 |
4. | 4 23 26 28 49 |
+---------------------------------------------+
Nick
[email protected]
Sean Lunde
by combine, I mean that my data is organized like this:
Month | Sales | Brand
1 | 25 | 1
2 | 13 | 1
3 | 19 | 1
4 | 23 | 1
1 | 21 | 2
2 | 17 | 2
3 | 19 | 2
4 | 26 | 2
1 | 11 | 3
2 | 19 | 3
3 | 19 | 3
4 | 28 | 3
What I want to do, is combine the values of sales when brand is equal
to 1 with the values when brand is equal to 2, creating a new brand,
with observations for each month. However, I want to keep the data
organized in the time periods.
On Oct 21, 2008, at 2:06 PM, Nick Cox wrote
> What does "combine" mean? If it means add, then know that
>
> gen z = x + y
>
> is the way to add two variables to produce a third.
Sean Lunde
> I am working with a set of panel data grouped by brand (the panel var)
> and date (the time var). I want to combine the values of some of the
> brand variables together,( e.g. combine the salesdollar value for
> brand x with salesdollar value for brand y). All of the variables I
> want to combine are numeric.
>
*
* 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/
