Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

RE: st: The function " invnorm(uniform())"


From   siyama@who.int
To   statalist@hsphsun2.harvard.edu
Subject   RE: st: The function " invnorm(uniform())"
Date   Thu, 24 Apr 2003 14:18:10 +0200

Many thanks to Nick Cox for his prompt reply - I will use the recommended
direct method.

Best regards,

Amani

-----Original Message-----
From: n j cox [mailto:n.j.cox@durham.ac.uk]
Sent: Thursday, 24 April 2003 13:47
To: statalist@hsphsun2.harvard.edu
Subject: Re: st: The function " invnorm(uniform())"


siyama@who.int wrote:
> 
> I am using the function "invnorm(uniform())" which returns normally
> distributed random numbers with mean        zero and standard deviation
one,
> inside a program "linsim.ado".    The latter is used with:
> 
> simul linsim, reps(10000)
> 
> program define lnsim
>             version 7.0
>             if "`1'"=="?" {
>                     global S_1 "eff diag1 diag2"
>                     exit
>             }
>             drop _all
>             set obs 1
>             gen eff = uniform()*100
>             gen diag1 = invnorm(uniform())
>             gen diag2 = invnorm(uniform())*1.5 + 5
>             post `1' (eff) (diag1) (diag2) 
> end
> 
> Can I say that the variable "diag2", is a normally distributed random
number
> with mean 5 and standard deviation 1.5.

Yes, but you need not use -simul- to get this. Just do it directly:

set obs 10000
gen eff = uniform()*100
gen diag1 = invnorm(uniform())
gen diag2 = invnorm(uniform())*1.5 + 5
outfile eff diag1 diag2

Nick
n.j.cox@durham.ac.uk





*
*   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/
*
*   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/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index