R devel - Dec 2009 - New version weighted mean differs from the old one (PR#14142)

Full_Name: Myung-Hoe Huh
Version: 2.10
OS: Windows
Submission from: (NULL) (116.120.84.194)


New Version (2.10.0) weighted mean produces unreasonable result: see below.

     wt <- c(5,  5,  4,  1)/15
     x <- c(3.7,3.3,3.5,2.8)
     x[4] <- NA
     (xm <- weighted.mean(x,wt,na.rm=T))

Outcome is 
>      (xm <- weighted.mean(x,wt,na.rm=T))
[1] 3.266667

The number is obtained by treating x[4] <- 0

I think the old version(2.8.0)'s weighte mean is more reasonable. The old
output
was
>      (xm <- weighted.mean(x,wt,na.rm=T))
[1] 3.5

The number si obtained ignoring the x[4], which is NA.

This was PR#14032. Fixed in R.10.1.


On Mon, 14 Dec 2009, huh420 at dreamwiz.com wrote:
> Full_Name: Myung-Hoe Huh
> Version: 2.10
> OS: Windows
> Submission from: (NULL) (116.120.84.194)
>
>
> New Version (2.10.0) weighted mean produces unreasonable result: see below.
>
>     wt <- c(5,  5,  4,  1)/15
>     x <- c(3.7,3.3,3.5,2.8)
>     x[4] <- NA
>     (xm <- weighted.mean(x,wt,na.rm=T))
>
> Outcome is
>
>>      (xm <- weighted.mean(x,wt,na.rm=T))
> [1] 3.266667
>
> The number is obtained by treating x[4] <- 0
>
> I think the old version(2.8.0)'s weighte mean is more reasonable. The
old output
> was
>
>>      (xm <- weighted.mean(x,wt,na.rm=T))
> [1] 3.5
>
> The number si obtained ignoring the x[4], which is NA.
>
> ______________________________________________
> R-devel at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-devel
>
Charles C. Berry                            (858) 534-2098
                                             Dept of Family/Preventive Medicine
E mailto:cberry at tajo.ucsd.edu	            UC San Diego
http://famprevmed.ucsd.edu/faculty/cberry/  La Jolla, San Diego 92093-0901