Re: [R] Ask for histogram

2010-01-13 Thread Yi Du
Thanks all, I fixed it. On Wed, Jan 13, 2010 at 2:47 PM, Don MacQueen wrote: > If I do > > b <- rnorm(4332) > hist(b,probability=T,breaks=30,col='lightblue',ylim=c(0,1)) > rug(b) > > The plot looks entirely reasonable. > > As far as being different from SAS, perhaps SAS and R use different

Re: [R] Ask for histogram

2010-01-13 Thread Don MacQueen
If I do b <- rnorm(4332) hist(b,probability=T,breaks=30,col='lightblue',ylim=c(0,1)) rug(b) The plot looks entirely reasonable. As far as being different from SAS, perhaps SAS and R use different breakpoints, that is, different boundaries between the histogram bars. -Don At 11:58 A

Re: [R] Ask for histogram

2010-01-13 Thread Steve Lianoglou
Hi, On Wed, Jan 13, 2010 at 12:58 PM, Yi Du wrote: > Hi, > > > I use a vector of data to draw the histogram, but it is different from the > graph by SAS. Can you check it for me please? How are we supposed to check something without data, pictures, etc? What do you want checking, exactly? > b i

[R] Ask for histogram

2010-01-13 Thread Yi Du
Hi, I use a vector of data to draw the histogram, but it is different from the graph by SAS. Can you check it for me please? b is a column vector of 4332 hist(b,probability=T,breaks=30,col='lightblue',ylim=c(0,1)) rug(b) When I used rug, I find the records are smaller than 4332. I don't know w