Re: st: -reshape long-

[David Kantor <kantor.d@att.net>]

At 01:20 AM 8/10/2008, Mandy Jia wrote:
> Hi all,
>
> I'm not familiar of the command -reshape-. So I was wondering if anyone
> could help me with an example of how to use -reshape- command.
>
> The data set I'm working on looks like this(w means wage)
> ------------------------------------------------------------------------
> -
> id age80 age81 age82 age93 age84  w80  w81  w82  w83  w84
>
> 1    18      19       20      21      22     10   11    13    14    14.2
> 2    19      20       21      22      23     15   16    16    14    15
> 3    17      18       19      20      21      9     9     10    11
> 11.4
> ---------------------------------------------------------------------
> When I reshape the data set, I want to limit the age to the range
> [19,20].
> In order to use fixed effects and random effects models to analyze, I
> want to reshape the data into:
> ---------------------------------------------------------------------
> id   age   w    year
> 1    19   11    81
> 1    20   13    82
> 2    19   15    80
> 2    20   16    81
> 3    19   10    82
> 3    20   11    83
> ---------------------------------------------------------------------
> What I tried is as follows:
>
> .reshape long age w ,i(id) j(year)
> .drop if age~=19 | age~20
> .sort id age
>
> My question is, is there any better way to do this? I'm not sure if
> what I did is OK, although I havn't found any weired thing in the
> result now.
Your command looks correct. The age filter is a separate matter (and 
I think you really meant -drop if age~=19 & age~=20).
You may want to inspect the data after the reshape -- before the age 
filter -- to convince yourself that it did what you wanted.

Also, I assume that in your data example you meant age83, rather than age93.

I can't comment on how this fits in fixed effects and random effects 
models. But in any case, it is good to be acquainted with -reshape-, 
and in most cases, a long shape is the most useful.

HTH.
--David

*
*   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/