David,
Thanks for taking the time to look into it. I tried your method and it seems
the "type" value wasn't really passed into my.boxplot.stats. If I manually
changed type value in my.boxplot.stats, I did see a difference. Attached
some data below. Thanks.
====================================================================
structure(list(ID = c(102, 103, 109, 112, 116, 120, 133, 137,
143, 148, 151, 154, 158, 164, 165, 170, 171, 172, 184, 188, 195,
196, 201, 205), CL = c(0.0278, 0.0265, 0.0198, 0.0112, 0.0297,
0.0251, 0.0282, 0.0219, 0.0372, 0.0244, 0.023, 0.0444, 0.0432,
0.0279, 0.0482, 0.051, 0.062, 0.0318, 0.0437, 0.0298, 0.0547,
0.135, 0.0417, 0.0286), V1 = c(2.99, 3.54, 3.06, 2.15, 2.85,
3.27, 2.71, 3.41, 2.87, 3.17, 2.83, 2.68, 2.51, 2.58, 2.15, 2.45,
2.43, 2.96, 2.41, 2.45, 2.12, 1.67, 2.23, 2.13), Q = c(0.185,
0.233, 0.225, 0.528, 0.194, 0.189, 0.0764, 0.109, 0.168, 0.219,
0.149, 0.144, 0.132, 0.107, 0.0389, 0.0954, 0.171, 0.138, 0.0863,
0.0512, 0.104, 0.088, 0.0863, 0.195), V2 = c(5.1, 4.69, 4.75,
3.45, 5.01, 5.06, 7.23, 6.34, 5.41, 4.9, 5.65, 5.72, 5.99, 6.48,
9.43, 6.76, 5.45, 5.9, 7.16, 8.69, 6.69, 7.11, 7.16, 5.29), ETA1 = c(0.192,
0.144, -0.146, -0.718, 0.257, 0.0902, -0.0989, -0.352, 0.179,
-0.243, -0.301, 0.355, 0.111, -0.326, 0.221, 0.278, 0.473, -0.195,
-0.182, -0.563, 0.0437, 0.947, -0.228, -0.606), ETA2 = c(-0.0444,
0.126, -0.0184, -0.374, -0.0892, 0.0467, -0.0213, 0.208, 0.0333,
0.133, 0.0205, -0.0338, -0.0148, 0.011, -0.168, -0.0406, -0.0475,
0.149, 0.0599, 0.079, -0.0644, -0.303, -0.018, -0.0613), ETA3 = c(-0.103,
0.127, 0.0931, 0.947, -0.0521, -0.0808, -0.701, -0.343, 0.0844,
0.353, -0.036, -0.0667, 0.0493, -0.165, -1.17, -0.277, 0.304,
0.0952, -0.0918, -0.614, 0.0941, -0.0717, -0.0921, 0.725), ETA4 = c(0.0378,
-0.0465, -0.0347, -0.353, 0.0188, 0.0302, 0.26, 0.129, -0.0308,
-0.13, 0.0135, 0.0248, -0.0184, 0.0606, 0.436, 0.102, -0.113,
-0.0343, 0.0338, 0.227, -0.0339, 0.0269, 0.0334, -0.27), DOSE = c(60,
60, 60, 60, 60, 60, 40, 40, 40, 40, 40, 40, 30, 30, 30, 30, 30,
30, 20, 20, 20, 20, 20, 20), FLAG = c(0L, 0L, 0L, 0L, 0L, 0L,
0L, 0L, 0L, 0L, 0L, 0L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L,
1L, 1L)), .Names = c("ID", "CL", "V1", "Q", "V2", "ETA1", "ETA2",
"ETA3", "ETA4", "DOSE", "FLAG"), row.names = c(1L, 7L, 12L, 15L,
21L, 27L, 33L, 39L, 45L, 48L, 54L, 60L, 65L, 67L, 71L, 76L, 81L,
86L, 91L, 93L, 98L, 103L, 108L, 112L), class = "data.frame")
On Fri, May 7, 2010 at 7:51 PM, David Winsemius <dwinsem...@comcast.net>wrote:
>
> On May 7, 2010, at 5:05 PM, David Winsemius wrote:
>
>
>> On May 7, 2010, at 4:18 PM, Jun Shen wrote:
>>
>> David,
>>>
>>> my.boxplot.stats is modified from boxplot.stats (package grDevices) as
>>> follows. x is the original argument, I guess it's the data object for
>>> processing. I only added "type". Thanks.
>>>
>>
>> Right, but you should be assigning stats=my.boxplot.stats, not
>> stats=myboxplot.stats(type=6). The functions then pick up their arguments
>> from their environment.
>>
>>
> Here's a worked example using modifications of a panel.bwplot help page
> call:
>
> (bwplot(voice.part ~ height, data = singer, stats=my.boxplot.stats,
> xlab = "Height (inches)",
> panel = function(...) {
> panel.grid(v = -1, h = 0)
> panel.bwplot(..., coef=3, type=1)
> },
> par.settings = list(plot.symbol = list(pch = 4))))
>
> One of the problems I faced was that I really could not see much difference
> between the results of changing the "type" for quantile. You provided no
> dataset for which you thought the difference made a .... difference. So I
> changed the coef argument to create an effect I could see.
>
>
>
> (And learn to send plain text. It is very easy to do so in gmail.)
>>
>> --
>> David.
>>
>>
>>> Jun
>>>
>>> =============================================================
>>> > boxplot.stats
>>>
>>> function (x, coef = 1.5, do.conf = TRUE, do.out = TRUE)
>>> {
>>> if (coef < 0)
>>> stop("'coef' must not be negative")
>>> nna <- !is.na(x)
>>> n <- sum(nna)
>>> stats <- stats::fivenum(x, na.rm = TRUE)
>>> iqr <- diff(stats[c(2, 4)])
>>> if (coef == 0)
>>> do.out <- FALSE
>>> else {
>>> out <- if (!is.na(iqr)) {
>>> x < (stats[2L] - coef * iqr) | x > (stats[4L] + coef *
>>> iqr)
>>> }
>>> else !is.finite(x)
>>> if (any(out[nna], na.rm = TRUE))
>>> stats[c(1, 5)] <- range(x[!out], na.rm = TRUE)
>>> }
>>> conf <- if (do.conf)
>>> stats[3L] + c(-1.58, 1.58) * iqr/sqrt(n)
>>> list(stats = stats, n = n, conf = conf, out = if (do.out) x[out &
>>> nna] else numeric(0L))
>>> }
>>> <environment: namespace:grDevices>
>>>
>>>
>>> On Fri, May 7, 2010 at 3:09 PM, David Winsemius <dwinsem...@comcast.net>
>>> wrote:
>>>
>>> On May 7, 2010, at 3:30 PM, Jun Shen wrote:
>>>
>>> Dear all,
>>>
>>> I constructed this function called my.boxplot.stats by replacing fivnum()
>>> with quantile() in function boxplot.stats(). So I can try different
>>> quantile
>>> methods in bwplot(). The problem is I couldn't pass different values to
>>> the
>>> "type" argument to my.boxplot.stats, which in turn is an argument in
>>> bwplot(). Now I just have to manually change the "type" value in
>>> my.boxplot.stats. What I would like to do is call bwplot() like
>>>
>>> bwplot(CL~DOSE,data=patab,stats=my.boxplot.stats(type=7)),
>>>
>>>
>>> You have defined a function with an argument of x that does not have a
>>> default value, so it throws an error when you call it with no value for x.
>>> (It's not complaining about the type argument.)
>>>
>>> --
>>> David.
>>>
>>>
>>> But I got an error:
>>> Error in my.boxplot.stats(type = 7) : element 1 is empty;
>>> the part of the args list of 'is.na' being evaluated was:
>>> (x)
>>>
>>> Thanks a lot.
>>>
>>> Jun Shen from Millipore Corporation
>>>
>>> ========================================================
>>> my.boxplot.stats<-function (x, coef = 1.5, *type=6*, do.conf = TRUE,
>>> do.out
>>> = TRUE)
>>> {
>>> if (coef < 0)
>>> stop("'coef' must not be negative")
>>> nna <- !is.na(x)
>>> n <- sum(nna)
>>> stats <- stats::quantile(x, type=type,na.rm = TRUE)
>>> iqr <- diff(stats[c(2, 4)])
>>> if (coef == 0)
>>> do.out <- FALSE
>>> else {
>>> out <- if (!is.na(iqr)) {
>>> x < (stats[2L] - coef * iqr) | x > (stats[4L] + coef *
>>> iqr)
>>> }
>>> else !is.finite(x)
>>> if (any(out[nna], na.rm = TRUE))
>>> stats[c(1, 5)] <- range(x[!out], na.rm = TRUE)
>>> }
>>> conf <- if (do.conf)
>>> stats[3L] + c(-1.58, 1.58) * iqr/sqrt(n)
>>> list(stats = stats, n = n, conf = conf, out = if (do.out) x[out &
>>> nna] else numeric(0L))
>>> }
>>>
>>> [[alternative HTML version deleted]]
>>>
>>> ______________________________________________
>>> R-help@r-project.org mailing list
>>> 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, MD
>>> West Hartford, CT
>>>
>>>
>>>
>> David Winsemius, MD
>> West Hartford, CT
>>
>> ______________________________________________
>> R-help@r-project.org mailing list
>> 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, MD
> West Hartford, CT
>
>
[[alternative HTML version deleted]]
______________________________________________
R-help@r-project.org mailing list
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.
