Taking Average by id

[Huseyin Unal]

Hi Everyone,
I have a dataset looking like this:

Date id var1 var2 var3 var4...
2010 1 ... ... ... ...
2010 1 ... ... ... ...
2010 1 ... ... ... ...
2010 2 ... ... ... ...
2010 2 ... ... ... ...
2010 2 ... ... ... ...
2010 3 ... ... ... ...
2010 3 ... ... ... ...
2010 3 ... ... ... ...

I'd like to find mean of each variable by id. That is, at the end, I want to achieve something like this:


Date id mean_var1 mean_var2 mean_var3 mean_var4...
2010 1 ... ... ... ...
2010 2 ... ... ... ...
2010 3 ... ... ... ...

It would be great if you can help. Thank you in advance.
Best,

[Felix Bittmann]

bysort id: egen meanvar = mean(var)
help egen
help collapse

[Nick Cox]

Felix Bittmann gives the flavour, but I guess that something like

by id Date : egen meanvar = mean(var)

may be closer to what is needed. The thread title and the desired example don't seem consistent, however.

[Carlo Lazzaro]

Huseyin:
most depends on what you're after, as the following toy-example try to highlight:

. use "http://www.stata-press.com/data/r15/nlswork.dta"
(National Longitudinal Survey.  Young Women 14-26 years of age in 1968)

. foreach var of varlist birth_yr-ln_wage {
  2. bysort idcode year: egen mean_`var'=mean(`var')
  3.  }

. list idcode year age mean_age in 1/3

     +--------------------------------+
     | idcode   year   age   mean_age |
     |--------------------------------|
  1. |      1     70    18         18 |
  2. |      1     71    19         19 |
  3. |      1     72    20         20 |
     +--------------------------------+

. foreach var of varlist birth_yr-ln_wage {
  2. bysort idcode: egen bis_mean_`var'=mean(`var')
  3.  }

. list idcode year age bis_mean_age in 1/3

     +--------------------------------+
     | idcode   year   age   bis~_age |
     |--------------------------------|
  1. |      1     70    18   26.33333 |
  2. |      1     71    19   26.33333 |
  3. |      1     72    20   26.33333 |
     +--------------------------------+

.

Obbviously, I would discard my first code, since (as expected), it gives back exactly the same value per id year of the original variable,