st: RE: Number of decimal points and rounding off when using esttab

["Hussein, Mustafa (Mustafa Hussien)" <mhussei4@uthsc.edu>]

Hi Amal,

when you use esttab, you could specify the number of digits beyond the decimal points in reported estimates by saying something like (specifying 3 figures):

esttab, b(3) ci(3) eform  // the no. of figures you specify or b usually carries over to its inferential statistics as well

And if you want to produce nice looking tables, you could specify other options as well, both in the esttab and the eststo steps.

On another tangent, if you are using all these if restrictions to get a consistent estimation sample, you could just make up a single dummy variable that summarizes all your conditions, or you could alternatively use if e(sample) for the rest of your models after model 1, if it works on the sample that you want. Having all these if restrictions increases the likelihood of errors in sample selection.

I hope that helps

Best,
Mustafa
________________________________________
From: owner-statalist@hsphsun2.harvard.edu [owner-statalist@hsphsun2.harvard.edu] on behalf of Amal Khanolkar [Amal.Khanolkar@ki.se]
Sent: Monday, October 28, 2013 9:05 AM
To: statalist@hsphsun2.harvard.edu
Subject: st: Number of decimal points and rounding off when using esttab

Hi all,

I use the following set of commands to get easy to read results from a set of regression models using the esttab package:

eststo clear
eststo: mlogit moth_wtgainbybmi i.ethnicity2 sex byear gestwk, base(1) rrr, if magecat2!=. & parity!=. & education2!=. & famsit_new2!=. & smoke2!=. & gestdt!=. & gestht!=. & MBMI5!=. & multibirth==1
eststo: mlogit moth_wtgainbybmi i.ethnicity2 sex byear gestwk ib2.magecat2 i.parity, base(1) rrr,  if education2!=. & famsit_new2!=. & smoke2!=. & gestdt!=. & gestht!=. & MBMI5!=. & multibirth==1
eststo: mlogit moth_wtgainbybmi i.ethnicity2 sex byear gestwk ib2.magecat2 i.parity ib2.education2 ib2.famsit_new2, base(1) rrr,  if smoke2!=. & gestdt!=. & gestht!=. & MBMI5!=. & multibirth==1
eststo: mlogit moth_wtgainbybmi i.ethnicity2 sex byear gestwk ib2.magecat2 i.parity ib2.education2 ib2.famsit_new2 ib2.smoke2, base(1) rrr,  if gestdt!=. & gestht!=. & MBMI5!=. & multibirth==1
eststo: mlogit moth_wtgainbybmi i.ethnicity2 sex byear gestwk ib2.magecat2 i.parity ib2.education2 ib2.famsit_new2 ib2.smoke2 i.gestdt i.gestht, base(1) rrr, if MBMI5!=. & multibirth==1
eststo: mlogit moth_wtgainbybmi i.ethnicity2 sex byear gestwk ib2.magecat2 i.parity ib2.education2 ib2.famsit_new2 ib2.smoke2 i.gestdt i.gestht ib2.MBMI5, base(1) rrr, if multibirth==1
esttab, eform ci


I know the options a1, a2, a3..... help specify the number of digits one would like after the decimal point. However, I wasn't able to figure out (and I might have missed this while reading the help file for esttab) how one can specify the number of digits after the decimal point (say if I wanted the estimate to have max two digits after the decimal point like 2.65, or 3.38) but these estimates are rounded off?


Thanks Amal.

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