[R] image function with date-time on X axis

Halldór Björnsson halldor.bjornsson at gmail.com
Fri Apr 30 11:24:04 CEST 2010


Thanks Peter

zlevs and ylevs are similar, that was not the problem

Following your suggestion I now use

> image(tax, ylevs, Rmat, xaxt = "n")
>  axis.POSIXct(1, at = seq(tax[1],tax[length(tax)],by="1 day"),format="%d-%m")

Which does what I wanted  to do
I still get the warning message, but otherwise this works.

Thanks again

Halldór


On Fri, Apr 30, 2010 at 2:05 AM, Peter Ehlers <ehlers at ucalgary.ca> wrote:
> On 2010-04-29 12:15, Halldór Björnsson wrote:
>>
>> I am trying to plot a image where the x axis has the units of time.
>> When I issue the
>> image(x,y,z) command with x as a POSIXct object, it fails to put a
>> time stamp on the
>> x axis.
>>
>> Instead I get a warning "Incompatible methods" warning and no dates on
>> my x axis.
>>
>> This example shows my problem:
>>
>> Rmat=t(matrix(data=rnorm(1:500),ncol=10,nrow=50))
>> tax=seq(ISOdate(2010,4,14,12,0,0), ISOdate(2010,4,19,00,0,0), by = "12
>> hours") ylevs=seq(100,5000,length=50)
>> image(tax,ylevs,Rmat)
>>
>> This givies the warning:
>> In image.default(tax, zlevs, Rmat) :
>>   Incompatible methods ("-.POSIXt", "Ops.difftime") for "-"
>> 2: In is.vector(X) :
>
> I don't get that warning. Are you doing something you're not
> telling us? (I note that you have 'ylevs' in one place and 'zlevs'
> in another).
>
> This works for me (using your above definitions of Rmat, etc):
>
>  image(tax, ylevs, Rmat, xaxt = "n")
>  axis(1, at = as.numeric(tax), lab = weekdays(tax, TRUE))
>
>  -Peter Ehlers
>
>>
>> Contrast the x axis result of image with that of plot
>>  plot(tax,rnorm(length(tax)))
>>
>> where the date-time shows stamps work fine.
>>
>> How can I get image to behave (or is there a better function for the job?)
>>
>> Sincerely
>> Halldór
>>
>> ______________________________________________
>> 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.
>>
>>
>
> --
> Peter Ehlers
> University of Calgary
>



More information about the R-help mailing list