Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: AW: Nº of Observations in Probit regressions


From   "Martin Weiss" <[email protected]>
To   <[email protected]>
Subject   st: AW: Nº of Observations in Probit regressions
Date   Thu, 5 Mar 2009 14:57:55 +0100

<> 

Just try it out by introducing missings into a panel

*************
webuse union, clear
probit union age grade not_smsa south southXt
loc full `e(N)'
foreach var of varlist  age grade not_smsa south southXt{
replace `var'=. if runiform()<0.02
}
probit union age grade not_smsa south southXt
di in red "Decline from " `full' " to " `e(N)' " due to missings"
*************



HTH
Martin


-----Ursprüngliche Nachricht-----
Von: [email protected]
[mailto:[email protected]] Im Auftrag von SMAC
Gesendet: Donnerstag, 5. März 2009 14:51
An: [email protected]
Betreff: st: Nº of Observations in Probit regressions 

Dear Statalisters,

I have a little doubt about the probit's output "Nº Obs."  what does it 
means exactly?
I have an unbalanced  panel with 5120 obs. (8 years x 640 firms in 
average), but when I run a probit regression (with probit command, not 
xtprobit)
the Nº obs specified in the output is 2225. What happens with the rest 
of observations? are they dropped due to missing values in independent 
variables?
Dependent Variable has not missing values, but among independents 
variables, there is several variables with missing data.
I would really appreciate your help,

Sergio


*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/


*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/



© Copyright 1996–2024 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index