R devel - May 2004 - Generating Lognormal Random variables (PR#6843)

Full_Name: Anthony Gichangi
Version: 1.90
OS: Windows XP Pro
Submission from: (NULL) (130.225.131.206)


The function rlnorm generates negative values for lognormal distribution. 

x- rlnorm(1000, meanlog =  0.6931472, sdlog = 1)

Regards

Anthony

anthony@stat.sdu.dk writes:
> Full_Name: Anthony Gichangi
> Version: 1.90
> OS: Windows XP Pro
> Submission from: (NULL) (130.225.131.206)
> 
> 
> The function rlnorm generates negative values for lognormal distribution. 
> 
> x- rlnorm(1000, meanlog =  0.6931472, sdlog = 1)
Eh?
> all(rlnorm(1000000, meanlog =  0.6931472, sdlog = 1) >= 0)
[1] TRUE

Try reproducing with a fixed random seed so that others have a chance
to check.

-- 
   O__  ---- Peter Dalgaard             Blegdamsvej 3  
  c/ /'_ --- Dept. of Biostatistics     2200 Cph. N   
 (*) \(*) -- University of Copenhagen   Denmark      Ph: (+45) 35327918
~~~~~~~~~~ - (p.dalgaard@biostat.ku.dk)             FAX: (+45) 35327907

On Sat,  1 May 2004 17:00:28 +0200 (CEST), anthony@stat.sdu.dk wrote:
>Full_Name: Anthony Gichangi
>Version: 1.90
>OS: Windows XP Pro
>Submission from: (NULL) (130.225.131.206)
>
>
>The function rlnorm generates negative values for lognormal distribution. 
>
>x- rlnorm(1000, meanlog =  0.6931472, sdlog = 1)
It doesn't for me, and the line you typed above makes it look like
you're subtracting lognormals from x.

Could you post a reproducible example?  For random number generators,
you can make them reproducible by using set.seed, e.g.
> set.seed(1)
> range(rlnorm(1000, meanlog =  0.6931472, sdlog = 1))
[1]  0.09877592 90.32586746

Duncan Murdoch