Plotting a count process

classic Classic list List threaded Threaded
4 messages Options
Reply | Threaded
Open this post in threaded view
|

Plotting a count process

voodooochild
hello everybody,

i want to plot a count process in the following way, but i don't know
how i can do that. i have data for example x<-(0,2,6,2,8,4,.....) and
dates y which is a vector of weekly dates for example
(01/01/06, 08/01/06, 15/01/06, 22/01/06, ....), now i want to plot the
y's an the horizontal axis. On each date the count process jumps upwards
1 unit, so the vertical axis is 0, 1, 2, 3, ......
the distance between the dates is shown in vector x, so for example the
distance between 08/01/06 and 15/01/06 should be 2. maybe i can use some
times series functions for doing this?

i would be very thankful for any advice.

best regards
Andreas

______________________________________________
[hidden email] mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Reply | Threaded
Open this post in threaded view
|

Re: Plotting a count process

Dieter Menne
 <voodooochild <at> gmx.de> writes:

> i want to plot a count process in the following way, but i don't know
> how i can do that. i have data for example x<-(0,2,6,2,8,4,.....) and
> dates y which is a vector of weekly dates for example
> (01/01/06, 08/01/06, 15/01/06, 22/01/06, ....), now i want to plot the
> y's an the horizontal axis. On each date the count process jumps upwards
> 1 unit, so the vertical axis is 0, 1, 2, 3, ......
> the distance between the dates is shown in vector x, so for example the
> distance between 08/01/06 and 15/01/06 should be 2. maybe i can use some
> times series functions for doing this?
>

If I understand you correctly, this means plotting equidistant point against
the cumulative sum. Forgetting about the dates now, this would do it.

x = floor(pmax(rnorm(30,5,2),1))
xcum = cumsum(x)
plot(xcum,1:30)

Dieter

______________________________________________
[hidden email] mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Reply | Threaded
Open this post in threaded view
|

Re: Plotting a count process

voodooochild
Thank you for your advice, but this is not the exaxt thing i'm looking for
http://robotics.caltech.edu/~zoran/Research/poisson/img1.png
this picture gives an example what i want to get. instead of t0, t1,
t2,.... i want to draw my dates.
In my description i forgot the the horizontal line, which indicates the
time between two dates each, sorry for that.

regards
andreas


Dieter Menne wrote:

> <voodooochild <at> gmx.de> writes:
>
>  
>
>>i want to plot a count process in the following way, but i don't know
>>how i can do that. i have data for example x<-(0,2,6,2,8,4,.....) and
>>dates y which is a vector of weekly dates for example
>>(01/01/06, 08/01/06, 15/01/06, 22/01/06, ....), now i want to plot the
>>y's an the horizontal axis. On each date the count process jumps upwards
>>1 unit, so the vertical axis is 0, 1, 2, 3, ......
>>the distance between the dates is shown in vector x, so for example the
>>distance between 08/01/06 and 15/01/06 should be 2. maybe i can use some
>>times series functions for doing this?
>>
>>    
>>
>
>If I understand you correctly, this means plotting equidistant point against
>the cumulative sum. Forgetting about the dates now, this would do it.
>
>x = floor(pmax(rnorm(30,5,2),1))
>xcum = cumsum(x)
>plot(xcum,1:30)
>
>Dieter
>
>______________________________________________
>[hidden email] mailing list
>https://stat.ethz.ch/mailman/listinfo/r-help
>PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
>
>
>  
>

______________________________________________
[hidden email] mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Reply | Threaded
Open this post in threaded view
|

Re: Plotting a count process

voodooochild
Now my problem is solved, i thinked a while about it and rearanged my
plot in the following way for everybody how is interested

x<-0:3
y<-c(0,5,8,3)
dates<-c("","01.01.05","08.01.05","15.01.05")
plot(x,cumsum(y),type="p",axes=FALSE,xlab="Date",ylab="Failure
Number",pch=20)

axis(side = 1,0:3 ,labels = dates)
axis(side = 2)
box()

i<-seq(along=x)
segments(x[i],cumsum(y)[i],x[i+1],cumsum(y)[i])
segments(x[length(x)],cumsum(y)[length(y)],x[length(x)]+1,cumsum(y)[length(y)])


regards
andreas


[hidden email] wrote:

>Thank you for your advice, but this is not the exaxt thing i'm looking for
>http://robotics.caltech.edu/~zoran/Research/poisson/img1.png
>this picture gives an example what i want to get. instead of t0, t1,
>t2,.... i want to draw my dates.
>In my description i forgot the the horizontal line, which indicates the
>time between two dates each, sorry for that.
>
>regards
>andreas
>
>
>Dieter Menne wrote:
>
>  
>
>><voodooochild <at> gmx.de> writes:
>>
>>
>>
>>    
>>
>>>i want to plot a count process in the following way, but i don't know
>>>how i can do that. i have data for example x<-(0,2,6,2,8,4,.....) and
>>>dates y which is a vector of weekly dates for example
>>>(01/01/06, 08/01/06, 15/01/06, 22/01/06, ....), now i want to plot the
>>>y's an the horizontal axis. On each date the count process jumps upwards
>>>1 unit, so the vertical axis is 0, 1, 2, 3, ......
>>>the distance between the dates is shown in vector x, so for example the
>>>distance between 08/01/06 and 15/01/06 should be 2. maybe i can use some
>>>times series functions for doing this?
>>>
>>>  
>>>
>>>      
>>>
>>If I understand you correctly, this means plotting equidistant point against
>>the cumulative sum. Forgetting about the dates now, this would do it.
>>
>>x = floor(pmax(rnorm(30,5,2),1))
>>xcum = cumsum(x)
>>plot(xcum,1:30)
>>
>>Dieter
>>
>>______________________________________________
>>[hidden email] mailing list
>>https://stat.ethz.ch/mailman/listinfo/r-help
>>PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
>>
>>
>>
>>
>>    
>>
>
>______________________________________________
>[hidden email] mailing list
>https://stat.ethz.ch/mailman/listinfo/r-help
>PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
>
>
>  
>

______________________________________________
[hidden email] mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html