How to combine multiple observations(in multiple variables) for an individual to one variable?

[Wolfram Ritter]

Hi Guys

I'm new here so please forgive me if I didn't create a perfect question.

My Data looks like this:

[CODE]
* Example generated by -dataex-. To install: ssc install dataex
clear
input float(o1 o2 o3 id Indep-Var1 Indep-Var2 )
1 0 1 1 1 0
0 1 1 2 1 0
0 1 1 3 1 0
0 1 1 4 0 1
0 0 1 5 0 1

In this example I have 3 observations per subject (o1 o2 o3) and I was wondering how to create a new variable oit (where i is the subject and t is the corresponding number of the observation) so i can create a regression that looks like this:

reg oit Indep-Var1 Indep-Var2,r


cheers

[Phil Bromiley]

I'm not sure exactly what you want. You can use bysort id: egen meanx=mean(x) to create means by id while retaining the original number of observations. You can use egen with the count function to give you number of observations per id, or egen with the total function to sum the observations.

Alternatively, if you want to move to one observation per subject you can use collapse.

[Rich Goldstein]

I also am not completely clear but I think (guess) you want -reshape-:

reshape long o, i(id) j(count)

or whatever you want your "j" variable to be called