Dear David:
Could you please check what I did to create a histogram for this data set.
Actually, I used barplot for myhist$counts. But I have two problems:
*1-* when I rebuild the x-axis, only the label of the first bar appear, but
not for the others.
*2-* I tried to add frequencies at top of bars, but I could not.
any helps will be highly appreciated
*First*, I replaced the detection limit values of 0.01 by a smaller value
of 0.009.
NH3Nconcentrations.hist<-c(0.009,0.009,0.009,0.009,0.009,0.009,0.009,0.009,0.009,0.009,0.009,0.009,0.009,0.009,0.009,0.009,
0.01,0.01,0.01,0.01,0.01,0.01,0.01,0.01,0.01,0.01,0.01,0.01,0.02,0.02,0.02,0.02,0.02,0.02,0.02,0.02,
0.02,0.02,0.02,0.02,0.02,0.02,0.03,0.03,0.03,0.03,0.03,0.03,0.03,0.04,0.04,0.04,0.04,0.04,0.04,0.05,
0.05,0.05,0.06,0.47)
myhist <- hist(NH3Nconcentrations.hist,
breaks=c(0.005,0.0099,0.01,0.02,0.03,0.04,0.05,0.06,0.40,0.50),
col=c(1,rep(2,8)), xaxt="n",ylim=c(0,20)) ### ,
*Second*, I used barplot as follows:
colors = c("red","gray","gray","gray","gray","gray","gray","gray","blue")
barplot(myhist$counts, space=0 ,ylim=c(0,20), col=colors)
####---rebuild the x-axis , But not work as it should be
axis(1, at=c(myhist$mids[1], myhist$breaks[-(1:2)]), labels=c("<0.01",
myhist$breaks[-(1:2)]))
with many thanks
steve
On Sat, Jan 2, 2016 at 11:38 AM, David Winsemius <dwinsem...@comcast.net>
wrote:
>
> > On Jan 2, 2016, at 2:24 AM, Steven Stoline <sstol...@gmail.com> wrote:
> >
> > Dear David:
> >
> > Thank you very much for the code, it works very good for this data set.
> >
> > I just have one more thing (if not bothered you).
> >
> > how about if some of the non-censored (fully measured) data equal to the
> detection limit?
> >
> > As an example, in the data set below, there are 16 censored observations
> with detection limit of 0.01, and there are some non-censored data
> observation equal to 0.01 (equal to the detection limit). I am wondering if
> we still can distinguish between them in the histogram. I tried to modify
> your code, but I could not make it work for this situation.
>
> I would probably construct an intermediate dataset copy where you
> "lowered" the items that were below the detection limit to a value ....
> below the detection limit, and then set the breaks parameter so that the
> real 0.01 items were included in the second bin.
>
> (That actually mimics what I usually do with the actual values in
> regression situations. I consider the measurements "below the detection
> limit" to still be meaningful.)
>
> --
> David.
> >
> > I crated a data frame, I want to create histogram for the variable
> "NH3Nconcentrations" (second column in the data frame).
> >
> >
> > Once again, thank you very much for your helps.
> >
> >
> >
> >
> >
> cen<-c(1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,
> > 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0)
> >
> >
> censored<-c(TRUE,TRUE,TRUE,TRUE,TRUE,TRUE,TRUE,TRUE,TRUE,TRUE,TRUE,TRUE,TRUE,TRUE,TRUE,TRUE,
> >
> FALSE,FALSE,FALSE,FALSE,FALSE,FALSE,FALSE,FALSE,FALSE,FALSE,FALSE,FALSE,FALSE,FALSE,FALSE,
> >
> FALSE,FALSE,FALSE,FALSE,FALSE,FALSE,FALSE,FALSE,FALSE,FALSE,FALSE,FALSE,FALSE,FALSE,FALSE,
> >
> FALSE,FALSE,FALSE,FALSE,FALSE,FALSE,FALSE,FALSE,FALSE,FALSE,FALSE,FALSE,FALSE,FALSE)
> >
> >
> data.original<-c("<0.01","<0.01","<0.01","<0.01","<0.01","<0.01","<0.01","<0.01","<0.01","<0.01",
> >
> "<0.01","<0.01","<0.01","<0.01","<0.01","<0.01",0.01,0.01,0.01,0.01,0.01,0.01,0.01,0.01,0.01,0.01,
> >
> 0.01,0.01,0.02,0.02,0.02,0.02,0.02,0.02,0.02,0.02,0.02,0.02,0.02,0.02,0.02,0.02,0.03,0.03,0.03,0.03,
> > 0.03,0.03,0.03,0.04,0.04,0.04,0.04,0.04,0.04,0.05,0.05,0.05,0.06,0.47)
> >
> >
> NH3Nconcentrations<-c(0.01,0.01,0.01,0.01,0.01,0.01,0.01,0.01,0.01,0.01,0.01,0.01,0.01,0.01,0.01,0.01
> >
> ,0.01,0.01,0.01,0.01,0.01,0.01,0.01,0.01,0.01,0.01,0.01,0.01,0.02,0.02,0.02,0.02,0.02,0.02,0.02,0.02,
> >
> 0.02,0.02,0.02,0.02,0.02,0.02,0.03,0.03,0.03,0.03,0.03,0.03,0.03,0.04,0.04,0.04,0.04,0.04,0.04,0.05,
> > 0.05,0.05,0.06,0.47)
> >
> >
> NH3N.concentrations<-data.frame(data.original,NH3Nconcentrations,cen,censored)
> >
> > attach(NH3N.concentrations)
> >
> >
> > NH3N.concentrations
> >
> >
> >
> > with many thanks
> > steve
> >
> > On Fri, Jan 1, 2016 at 3:42 PM, David Winsemius <dwinsem...@comcast.net>
> wrote:
> >
> >> On Jan 1, 2016, at 3:45 AM, Steven Stoline <sstol...@gmail.com> wrote:
> >>
> >> Dear Rolf:
> >>
> >>
> >> The histogram should contain a bar(s) for the censored data values
> replaced
> >> by their detection limit(s) with different color than other bars for the
> >> noncensored values . In this example there are only 3 censored values
> with
> >> only one detection limit of DL = 1450.
> >>
> >>
> >> with many thanks
> >> steve
> >>
> >>
> >>
> >> On Thu, Dec 31, 2015 at 4:16 PM, Rolf Turner <r.tur...@auckland.ac.nz>
> >> wrote:
> >>
> >>> On 31/12/15 23:20, Steven Stoline wrote:
> >>>
> >>>> Dear All:
> >>>>
> >>>> I need helps with creating histograms for data that include left
> >>>> censored observations.
> >>>>
> >>>> Here is an example of left censored data
> >>>>
> >>>>
> >>>>
> >>>> *Sulfate.Concentration*
> >>>> <-matrix(c(1450,1800,1840,1820,1860,1780,1760,1800,1900,1770,1790,
> >>>> 1780,1850,1760,1450,1710,1575,1475,1780,1790,1780,1450,1790,1800,
> >>>> 1,0,0,0,0,0,0,0,0,0,0,0,0,0,1,0,0,0,0,0,0,1,0,0),24,2)
> >>>>
> >
> > myhist <- hist(sulfate[,1],
> breaks=c(1400,1451,1500,1600,1700,1800,1900), col=c(1,rep(2,5)), xaxt="n")
> > # plots with no x axis labeling
> > myhist
> > #------------------
> > $breaks
> > [1] 1400 1451 1500 1600 1700 1800 1900
> >
> > $counts
> > [1] 3 1 1 0 14 5
> >
> > $density
> > [1] 0.0024509804 0.0008503401 0.0004166667 0.0000000000 0.0058333333
> 0.0020833333
> >
> > $mids
> > [1] 1425.5 1475.5 1550.0 1650.0 1750.0 1850.0
> >
> > $xname
> > [1] "sulfate[, 1]"
> >
> > $equidist
> > [1] FALSE
> >
> > attr(,"class")
> > [1] "histogram"
> > #---rebuild the x-axis ----------------
> > axis(1, at=c(myhist$mids[1],myhist$breaks[-(1:2)]), labels=c("<1450",
> myhist$breaks[-(1:2)]))
> >
> > <Rplot001.png>
> >
> > --
> > David.
> >
> >>>>
> >>>> *Column 2* is an indicator for censoring "*1*" for left censored
> >>>> observations and "*0*" for non-censored (fully measured)
> >>>> observations.
> >>>>
> >>>
> >>> And what, pray tell, do you want the resulting histogram to look like?
> >>> See e.g. fortune("mind_read").
> >>>
> >>> cheers,
> >>>
> >>> Rolf Turner
> >>>
> >>> --
> >>> Technical Editor ANZJS
> >>> Department of Statistics
> >>> University of Auckland
> >>> Phone: +64-9-373-7599 ext. 88276
> >>>
> >>
> >>
> >>
> >> --
> >> Steven M. Stoline
> >> 1123 Forest Avenue
> >> Portland, ME 04112
> >> sstol...@gmail.com
> >>
> >> [[alternative HTML version deleted]]
> >>
> >> ______________________________________________
> >> R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see
> >> https://stat.ethz.ch/mailman/listinfo/r-help
> >> PLEASE do read the posting guide
> http://www.R-project.org/posting-guide.html
> >> and provide commented, minimal, self-contained, reproducible code.
> >
> > David Winsemius
> > Alameda, CA, USA
> >
> >
> >
> >
> > --
> > Steven M. Stoline
> > 1123 Forest Avenue
> > Portland, ME 04112
> > sstol...@gmail.com
>
> David Winsemius
> Alameda, CA, USA
>
>
--
Steven M. Stoline
1123 Forest Avenue
Portland, ME 04112
sstol...@gmail.com
[[alternative HTML version deleted]]
______________________________________________
R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
