estat sum error 498

[Christopher Weber]

Hello everyone,

I want to generate descriptive statistics after running my regression, using the "estat sum" command. However, if i try to run "estat sum", i get the following error code: ln_xrdintensity has 1 missing observation r(498). I do not understand the problem as the regression is running well. Can someone help me?


regression code:

xtpoisson patents cl.ln_xrdintensity##cl.Numberofmergers cl2.ln_xrdintensity##cl2.Numberofmergers cl3.ln_xrdintensity##cl3.Numberofmergers l.ln_emp i.YearEffective i. AcquirorPrimarySICCode, fe vce (robust)

estat code I want to run afterwards:

estat sum

best regards

Chris

[William Lisowski]

That seems like a bug to me. The essence of estat sum is that it calculates summary statistics on the estimation sample - those observations actually used by the model. I assume you have run the estat sum immediately following the xtpoisson with no intervening commands that might have an unintended side effect.

I was able to create a simple reproducible example of the problem.

clear all
cls
webuse nlswork, clear
xtset idcode year
xtreg ln_wage l.tenure
estat sum
estat sum l.tenure
summarize l.tenure if e(sample)

. webuse nlswork, clear
(National Longitudinal Survey of Young Women, 14-24 years old in 1968)

. xtset idcode year

Panel variable: idcode (unbalanced)
 Time variable: year, 68 to 88, but with gaps
         Delta: 1 unit

. xtreg ln_wage l.tenure

Random-effects GLS regression                   Number of obs     =     10,676
Group variable: idcode                          Number of groups  =      3,677
...

. estat sum
tenure has 148 missing observations
r(498);

. estat sum l.tenure
tenure has 148 missing observations
r(498);

. summarize l.tenure if e(sample)

    Variable |        Obs        Mean    Std. dev.       Min        Max
-------------+---------------------------------------------------------
      tenure |
         L1. |     10,676    2.722501    3.457858          0       24.5

.

This suggests a workaround to solve your immediate problem.

[William Lisowski]

I forwarded the above problem to Stata Technical Services and received a reply apologizing for the inconvenience and saying that it has been forwarded to the developers to be fixed in a future update.

[Christopher Weber]

Thank you very much William! That helped me a lot