> 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
______________________________________________
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.
