Hello all,

I ran this simple command below to create a table of summary statistics. My data set has 284,910 observations.
However, e(sum) yields 5,397,544 observation in total (see "age" row). That number is the raw data total--the majority of these observations were dropped.
How does Stata "remember" this e(sum) total? Does anyone have an idea how I get the correct number (284,910) of observations for the cell "age" / e(sum)?
Note that the e(sum) values for "white," "black," and so forth are correct--adding up to 284,910.

Code:
estpost summarize $sum_vars, listwise

| e(count) e(sum_w) e(mean) e(Var) e(sd) e(min) e(max) e(sum)
-------------+----------------------------------------------------------------------------------------
age | 284910 284910 18.94473 3.80579 1.950843 16 22 5397544
white | 284910 284910 .7102067 .2058139 .4536671 0 1 202345
black | 284910 284910 .1315854 .1142711 .3380401 0 1 37490
asian | 284910 284910 .0535643 .0506953 .2251562 0 1 15261
other_race | 284910 284910 .1046436 .0936936 .3060941 0 1 29814
...

Code:
sum age

    Variable |        Obs        Mean    Std. Dev.       Min        Max
-------------+---------------------------------------------------------
         age |    284,910    18.94473    1.950843         16         22
I do run
Code:
estimates drop _all
and
Code:
eststo clear
at the start.


Thank you!