Assuming this is exp(sd(log())), I guess most
people would take the logs, etc. Writing your
own is not too difficult; below is a sketch.
------------------------------ begin geomeansd
program geomeansd
version 8
syntax [varlist] [if] [in]
marksample touse, novarlist
qui count if `touse'
if r(N) == 0 error 2000
if r(N) == 1 error 2001
local shown = 0
qui foreach v of local varlist {
capture confirm string var `v'
if _rc {
if !`shown' {
noi Displayheader
local shown = 1
}
tempvar logv
gen `logv' = log(`v') if `touse'
su `logv' if `touse'
noi di as txt abbrev("`v'",12) _col(15) ///
as res %6.0f r(N) ///
as res %15.3f exp(r(mean)) ///
as res %15.3f exp(r(sd))
drop `logv'
}
}
end
program Displayheader
di
di as txt "{space 17}Obs Geom. Mean Geom. SD"
di "{hline 50}"
end
------------------------ end geomeansd.ado
Nick
[email protected]
> I have a set of data that follows log-normal distribution.
> In STATA the
> command "means" only gives geometric mean and confidence
> intervals. Is there
> a way to calculate geometric standard deviation in STATA?
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
