<>
Assuming you want the upper p-val, you can of course do this by hand, as
seen in this example:
**************
clear*
set obs 1000
//id
gen id=_n
lab var id "Person ID"
//individual effect
gen alpha=rnormal(0,0.01)
label var alpha "Individual Effect"
//expand to # of time periods
expand 10
//get time period (after expansion!)
bys id: gen time=_n
lab var time "Time Period"
//residual per time period
gen eps=rnormal()
lab var eps "Idiosyncratic Shock"
//covariates
gen x1=rnormal()
gen x2=rnormal()
gen x3=rnormal()
//DGP
gen y=1+alpha+eps
compress
//xtset the thing
xtset id time
xtreg y x1 x2 x3, fe
local p= Ftail( e(df_b), e(df_r), e(F))
outreg2 using myfile.txt, /*
*/ adds("F test: ", e(F), Prob > F, `p') replace
seeout using "myfile.txt"
**************
HTH
Martin
-----Original Message-----
From: [email protected]
[mailto:[email protected]] On Behalf Of
[email protected]
Sent: Sonntag, 27. September 2009 12:00
To: [email protected]
Subject: st: Different p-values in outreg2 and regression output
Hello all,
I'm using -outreg2- to export panel regression results in excel. I
would like to have the p-values corresponding to each F-stat of each
regression (-xtreg, fe-) as well as some other statistics. Thus, I
wrote:
outreg2 [M1, M2, M3] using myfile.xls, e(N, r2_a, r2_w, r2_b, r2_o F, p)
Looking at the output, I found that the p-values listed in the
exported file are not exactly the same as those directly in the
regression. For example, in the Stata-regression output Prob>F=0.0022
and in the exported file it is 0.00264. Yet, the F-stats are correct.
Is it just a strange rounding convention or is there a logical
explanation for this?
Thanks for any help!
Ida
*
* 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/
