Regression tables with esttab - display confidence interval & p-value

[Megan Moreton]

Hi,

I am using esttab to create tables after logistic regression. I present the OR & the 95%CI - is it possible to present the p-values in the same table (eg, in the next column).

For example, if I run the following, could I also present the p-values in the table from the esttab command:

eststo clear
sysuse auto, clear

eststo: logistic foreign price mpg
eststo: logistic foreign price mpg headroom weight

esttab, b(%8.2f) eform ci label nonumbers mtitles("M1" "M2")

Thanks for any help,
Megan

[Andrew Musau]

esttab is from Stata Journal, as you are asked to explain in FAQ Advice #12.

eststo clear
sysuse auto, clear
eststo: logistic foreign price mpg
eststo: logistic foreign price mpg headroom weight
esttab, cells(b(fmt(3)) ci(fmt(3) par) p(fmt(3) par)) eform ///
gaps lines collabels(none) label nonumbers mtitles("M1" "M2") ///
addnotes("95% confidence intervals in brackets; p-values in parentheses.")

Res.:

. esttab, cells(b(fmt(3)) ci(fmt(3) par) p(fmt(3) par)) eform ///
> gaps lines collabels(none) label nonumbers mtitles("M1" "M2") ///
> addnotes("95% confidence intervals in brackets; p-values in parentheses.")

----------------------------------------------
                               M1           M2
----------------------------------------------
Car type                                      
Price                       1.000        1.001
                     [1.000,1.000] [1.000,1.002]
                          (0.022)      (0.003)

Mileage (mpg)               1.263        0.885
                     [1.108,1.441] [0.733,1.069]
                          (0.000)      (0.204)

Headroom (in.)                           1.173
                                  [0.298,4.612]
                                       (0.819)

Weight (lbs.)                            0.993
                                  [0.989,0.997]
                                       (0.001)
----------------------------------------------
Observations                   74           74
----------------------------------------------
Exponentiated coefficients
95% confidence intervals in brackets; p-values in parentheses.

[Megan Moreton]

very helpful - thank you

[Katherine Oleas]

Good morning, I have a question and if I don't want the p-value to come out, if not the stars, how do I do it?

[Andrew Musau]

Drop -p()- and add the option -star- within -b()-. And for us in economics, we want significance levels 0.01, 0.05 and 0.1.

eststo clear
sysuse auto, clear
eststo: logistic foreign price mpg
eststo: logistic foreign price mpg headroom weight
esttab, cells(b(fmt(3) star) ci(fmt(3) par)) eform ///
gaps lines collabels(none) label nonumbers ///
mtitles("M1" "M2") starlevels(* 0.1 ** 0.05 *** 0.01) ///
addnotes("95% confidence intervals in brackets." ///
"* p<0.1, ** p<0.05, *** p<0.01")

Res.:

. esttab, cells(b(fmt(3) star) ci(fmt(3) par)) eform ///
> gaps lines collabels(none) label nonumbers ///
> mtitles("M1" "M2") starlevels(* 0.1 ** 0.05 *** 0.01) ///
> addnotes("95% confidence intervals in brackets." ///
> "* p<0.1, ** p<0.05, *** p<0.01")

----------------------------------------------------
                               M1              M2   
----------------------------------------------------
Car type                                            
Price                       1.000**         1.001***
                     [1.000,1.000]    [1.000,1.002]   

Mileage (mpg)               1.263***        0.885   
                     [1.108,1.441]    [0.733,1.069]   

Headroom (in.)                              1.173   
                                     [0.298,4.612]   

Weight (lbs.)                               0.993***
                                     [0.989,0.997]   
----------------------------------------------------
Observations                   74              74   
----------------------------------------------------
Exponentiated coefficients
95% confidence intervals in brackets.
* p<0.1, ** p<0.05, *** p<0.01

[Katherine Oleas]

Thank you so much!!!

[Mahmoud Amoli]

Drop -p()- and add the option -star- within -b()-. And for us in economics, we want significance levels 0.01, 0.05 and 0.1.

eststo clear
sysuse auto, clear
eststo: logistic foreign price mpg
eststo: logistic foreign price mpg headroom weight
esttab, cells(b(fmt(3) star) ci(fmt(3) par)) eform ///
gaps lines collabels(none) label nonumbers ///
mtitles("M1" "M2") starlevels(* 0.1 ** 0.05 *** 0.01) ///
addnotes("95% confidence intervals in brackets." ///
"* p<0.1, ** p<0.05, *** p<0.01")

Res.:

. esttab, cells(b(fmt(3) star) ci(fmt(3) par)) eform ///
> gaps lines collabels(none) label nonumbers ///
> mtitles("M1" "M2") starlevels(* 0.1 ** 0.05 *** 0.01) ///
> addnotes("95% confidence intervals in brackets." ///
> "* p<0.1, ** p<0.05, *** p<0.01")

----------------------------------------------------
M1 M2
----------------------------------------------------
Car type
Price 1.000** 1.001***
[1.000,1.000] [1.000,1.002]

Mileage (mpg) 1.263*** 0.885
[1.108,1.441] [0.733,1.069]

Headroom (in.) 1.173
[0.298,4.612]

Weight (lbs.) 0.993***
[0.989,0.997]
----------------------------------------------------
Observations 74 74
----------------------------------------------------
Exponentiated coefficients
95% confidence intervals in brackets.
* p<0.1, ** p<0.05, *** p<0.01

Hello, I have an additional question, and I need your help: how may I get my results in the following format in columns where column 1 is the coefficient, the second column is the confidence interval in parenthesis (1.23,1.47), and column 3 is the p-value. I appreciate your help.

estimates	confidence interval	pvalue
1.24	(1.23,1.47)	0.023

[Andrew Musau]

Look at the placement of the double quotes within -cells()-.

eststo clear
sysuse auto, clear
eststo: logistic foreign price mpg headroom weight
esttab, cells("b(fmt(3)) ci(fmt(3) par) p(fmt(3))") eform ///
gaps lines collabels(OR CI p) label nonumbers ///
nomtitles starlevels(* 0.1 ** 0.05 *** 0.01) ///
addnotes("* p<0.1, ** p<0.05, *** p<0.01") compress

Res.:

. esttab, cells("b(fmt(3)) ci(fmt(3) par) p(fmt(3))") eform ///
> gaps lines collabels(OR CI p) label nonumbers ///
> nomtitles starlevels(* 0.1 ** 0.05 *** 0.01) ///
> addnotes("* p<0.1, ** p<0.05, *** p<0.01") compress

----------------------------------------------
                        OR        CI         p
----------------------------------------------
Car type                                      
Price                1.001 [1.000,1.002]     0.003

Mileage (mpg)        0.885 [0.733,1.069]     0.204

Headroom (in.)       1.173 [0.298,4.612]     0.819

Weight (lbs.)        0.993 [0.989,0.997]     0.001
----------------------------------------------
Observations            74                    
----------------------------------------------
Exponentiated coefficients
* p<0.1, ** p<0.05, *** p<0.01

[Mahmoud Amoli]

thank you so much, appreciate it.