Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at

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

st: AW: ratio function

From   "Martin Weiss" <>
To   <>
Subject   st: AW: ratio function
Date   Thu, 1 Apr 2010 14:04:54 +0200


So you want the ratio of means and se, but that seems to be exactly what
-ratio- gives you:

webuse highschool, clear

svy: ratio height/weight

quietly svy: mean height weight
nlcom (myratio: _b[height]/_b[weight])


-----Ursprüngliche Nachricht-----
[] Im Auftrag von Roman Kasal
Gesendet: Donnerstag, 1. April 2010 13:58
Betreff: st: ratio function

dear Sir/Madam

How to come around command like this?

svy: mean(columnA)/mean(columnB)

I would like to estimate ratio of means and its standard error. How is
it possible with STATA?
It is strange that function "ratio" takes into account just pairs....but
why when "ratio" divides whole mean(columnA)/mean(columnB)? Why the
function ignores missing values? In real life is very improbable that I
have perfect paired variables.

thank you

Roman Kasal

*   For searches and help try:

*   For searches and help try:

© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index