thr3ads.net - R help - [R] Plotting longitudinal data [Sep 2010]

If this information is useful, please help other people find it:
Share via:

Jukka Koskela

2010-Sep-07 16:27 UTC

[R] Plotting longitudinal data

Hello,

Hope that someone could help me plotting longitudinal data below:

7213	3333330001	0.8300	13.05.09	1
1	3333330001	0.8700	09.02.05	NULL
4797	3333330001	0.7700	21.03.07	NULL
2399	3333330001	0.7800	12.04.06	NULL
2400	3333330002	NULL	27.03.06	NULL
7230	3333330002	0.8200	14.05.09	0
2	3333330002	0.8400	09.02.05	NULL
4798	3333330002	0.8700	20.03.07	0
4799	3333330003	0.9000	20.03.07	13
2401	3333330003	0.9300	27.03.06	16
3	3333330003	0.8400	10.02.05	NULL
7233	3333330003	NULL	14.05.09	1
4	3333330004	0.7200	10.02.05	NULL
4800	3333330004	0.8900	19.03.07	22
2402	3333330004	0.7300	29.03.06	27
7258	3333330004	0.7700	18.05.09	1

The second column is a patient_id, the third is the value I want to  
plot against the fourth which is the date.

First I 'aggregate' the patient_ids:

id<-unique(dat$patient_id)

Then I try (and fail) to create a loop, that is supposed to plot the data:

for(i in 1:8480){patient_id==id[i]plot(date,value)}


What might be wrong?

And how could I only plot eg quintiles for the ones that go down fastest?

Thanks,

Jukka

Dennis Murphy

2010-Sep-07 17:00 UTC

head link

[R] Plotting longitudinal data

Hi:

There are several problems; see inline.

On Tue, Sep 7, 2010 at 9:27 AM, Jukka Koskela
<jukka.koskela@helsinki.fi>wrote:
> Hello,
>
> Hope that someone could help me plotting longitudinal data below:
>
Firstly, you want to use NA in place of NULL as the missing value code. This
is easy to change in a text editor.
>
> 7213    3333330001      0.8300  13.05.09        1
> 1       3333330001      0.8700  09.02.05        NULL
> 4797    3333330001      0.7700  21.03.07        NULL
> 2399    3333330001      0.7800  12.04.06        NULL
> 2400    3333330002      NULL    27.03.06        NULL
> 7230    3333330002      0.8200  14.05.09        0
> 2       3333330002      0.8400  09.02.05        NULL
> 4798    3333330002      0.8700  20.03.07        0
> 4799    3333330003      0.9000  20.03.07        13
> 2401    3333330003      0.9300  27.03.06        16
> 3       3333330003      0.8400  10.02.05        NULL
> 7233    3333330003      NULL    14.05.09        1
> 4       3333330004      0.7200  10.02.05        NULL
> 4800    3333330004      0.8900  19.03.07        22
> 2402    3333330004      0.7300  29.03.06        27
> 7258    3333330004      0.7700  18.05.09        1
>
> The second column is a patient_id, the third is the value I want to plot
> against the fourth which is the date.
>
This isn't hard to do. Below is one approach using package ggplot2, but
there are other ways.


mydat <- read.table(textConnection("
+ 7213    3333330001      0.8300  13.05.09        1
+ 1       3333330001      0.8700  09.02.05        NA
<snip for brevity>
+ 7258    3333330004      0.7700  18.05.09        1"), header = FALSE)
closeAllConnections()
# Convert the date to a Date object
mydat$date <- as.Date(mydat$V4, format = '%d.%m.%y')
mydat
     V1         V2   V3       V4 V5       date
1  7213 3333330001 0.83 13.05.09  1 2009-05-13
2     1 3333330001 0.87 09.02.05 NA 2005-02-09
3  4797 3333330001 0.77 21.03.07 NA 2007-03-21
4  2399 3333330001 0.78 12.04.06 NA 2006-04-12
5  2400 3333330002   NA 27.03.06 NA 2006-03-27
...

library(ggplot2)
g <- ggplot(mydat2, aes(x = date, y = V3, group = V2))
g + geom_line(aes(colour = V2), size = 1) +
      scale_colour_gradient(legend = FALSE)

This gives you a plot of V3 by Date for each patient; the lines are broken
when NAs are present in either y or x.


First I 'aggregate' the patient_ids:>
> id<-unique(dat$patient_id)
>
> Then I try (and fail) to create a loop, that is supposed to plot the data:
>
The loop below isn't going to work. Are you trying to produce 8480 separate
plots by patient or are you trying to put all 8480 in the same panel? If the
latter, I'd strongly encourage you to read about alpha transparency.
>
> for(i in 1:8480){patient_id==id[i]plot(date,value)}
>
>
> What might be wrong?
>
> I don't understand what you're driving at in the statement below -
perhapsyou could explain what you mean?

> And how could I only plot eg quintiles for the ones that go down fastest?
>

HTH,
Dennis

>
> Thanks,
>
> Jukka
>
> ______________________________________________
> 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.
>
	[[alternative HTML version deleted]]

Joshua Wiley

2010-Sep-07 17:11 UTC

head link

[R] Plotting longitudinal data

Hi,

Reading in (and slightly editing your data, so that patient_id is a
factor, and the date is class date).  Results of dput() provided for
others' benefit.

dat <- structure(list(V1 = c("7213", "1",
"4797", "2399", "2400", "7230",
"2", "4798", "4799", "2401",
"3", "7233", "4", "4800",
"2402",
"7258"), patient_id = structure(c(1L, 1L, 1L, 1L, 2L, 2L, 2L,
2L, 3L, 3L, 3L, 3L, 4L, 4L, 4L, 4L), .Label = c("3333330001",
"3333330002", "3333330003", "3333330004"), class =
"factor"),
    value = c(0.83, 0.87, 0.77, 0.78, NA, 0.82, 0.84, 0.87, 0.9,
    0.93, 0.84, NA, 0.72, 0.89, 0.73, 0.77), date = structure(c(14377,
    12823, 13593, 13250, 13234, 14378, 12823, 13592, 13592, 13234,
    12824, 14378, 12824, 13591, 13236, 14382), class = "Date"),
    V5 = c("1", "NULL", "NULL", "NULL",
"NULL", "0", "NULL",
    "0", "13", "16", "NULL",
"1", "NULL", "22", "27",
"1")), .Names = c("V1",
"patient_id", "value", "date", "V5"),
row.names = c(NA, -16L), class "data.frame")

Here are two simple little plots:

library(ggplot2)
ggplot(data = dat, aes(x = date, y = value, colour = patient_id)) + geom_line()

library(lattice)
xyplot(value ~ date, data = dat, groups = patient_id, type = "l")

Cheers,

Josh



On Tue, Sep 7, 2010 at 9:27 AM, Jukka Koskela <jukka.koskela at
helsinki.fi> wrote:> Hello,
>
> Hope that someone could help me plotting longitudinal data below:
>
> 7213 ? ?3333330001 ? ? ?0.8300 ?13.05.09 ? ? ? ?1
> 1 ? ? ? 3333330001 ? ? ?0.8700 ?09.02.05 ? ? ? ?NULL
> 4797 ? ?3333330001 ? ? ?0.7700 ?21.03.07 ? ? ? ?NULL
> 2399 ? ?3333330001 ? ? ?0.7800 ?12.04.06 ? ? ? ?NULL
> 2400 ? ?3333330002 ? ? ?NULL ? ?27.03.06 ? ? ? ?NULL
> 7230 ? ?3333330002 ? ? ?0.8200 ?14.05.09 ? ? ? ?0
> 2 ? ? ? 3333330002 ? ? ?0.8400 ?09.02.05 ? ? ? ?NULL
> 4798 ? ?3333330002 ? ? ?0.8700 ?20.03.07 ? ? ? ?0
> 4799 ? ?3333330003 ? ? ?0.9000 ?20.03.07 ? ? ? ?13
> 2401 ? ?3333330003 ? ? ?0.9300 ?27.03.06 ? ? ? ?16
> 3 ? ? ? 3333330003 ? ? ?0.8400 ?10.02.05 ? ? ? ?NULL
> 7233 ? ?3333330003 ? ? ?NULL ? ?14.05.09 ? ? ? ?1
> 4 ? ? ? 3333330004 ? ? ?0.7200 ?10.02.05 ? ? ? ?NULL
> 4800 ? ?3333330004 ? ? ?0.8900 ?19.03.07 ? ? ? ?22
> 2402 ? ?3333330004 ? ? ?0.7300 ?29.03.06 ? ? ? ?27
> 7258 ? ?3333330004 ? ? ?0.7700 ?18.05.09 ? ? ? ?1
>
> The second column is a patient_id, the third is the value I want to plot
> against the fourth which is the date.
>
> First I 'aggregate' the patient_ids:
>
> id<-unique(dat$patient_id)
>
> Then I try (and fail) to create a loop, that is supposed to plot the data:
>
> for(i in 1:8480){patient_id==id[i]plot(date,value)}
>
>
> What might be wrong?
>
> And how could I only plot eg quintiles for the ones that go down fastest?
>
> Thanks,
>
> Jukka
>
> ______________________________________________
> R-help at 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.
>


-- 
Joshua Wiley
Ph.D. Student, Health Psychology
University of California, Los Angeles
http://www.joshuawiley.com/

Seemingly Similar Threads

Search for more seemingly similar threads

R help - Sep 2010 - Plotting longitudinal data

[R] Plotting longitudinal data

[R] Plotting longitudinal data

[R] Plotting longitudinal data

Seemingly Similar Threads