reshape

[Nader Mehri]

Hi everyone,
I have a long dataset and used this code to reshape it to wide format:

ds hhidpn wave, not
reshape wide `r(varlist)', i(hhidpn) j(wave)

I was wondering to know how can I reshape the data to long format once again. I tried the below code, but it did not work:

reshape long `r(varlist)' , i(hhidpn) j(wave)

Thanks.
Nader

[Clyde Schechter]

After the first -reshape wide- runs, the `r(varlist)' disappears, because -reshape wide- itself leaves behind things in r(), so the previous contents of r() are gone.

Here's what you can do:

ds hhidpn wave, not
local reshape_vars `r(varlist)'

reshape wide `reshape_vars', i(hhidpn) j(wave)

reshape long `reshape_vars', i(hhidpn) j(wave)

By preserving the contents of `r(varlist)' in a local macro, you can resuse it
Actually, you don't even have to go to all that trouble. If you don't do anything between the two -reshape-s that disturbs the organization of the data, the second command can just simply be -reshape long- and Stata will remember what the variables, i(), and j() from the previous reshape were and use them.

[Nader Mehri]

Thanks. I got an error when trying to reshape from wide to long:
reshape long `reshape_vars', i(hhidpn) j(wave) invalid syntax r(198);

[Clyde Schechter]

My best guess is that you did not run the code all continuously from the point where local reshape_vars is defined and the point where you tried to -reshape long-. If there was any interruption in the code between those points, then -reshape_vars- would no longer exist and that would cause that problem.