[R] Zoo Data

arun smartpink111 at yahoo.com
Sat Mar 9 18:59:03 CET 2013


HI Jakob,

If your data is based on 30 min interval, this should work:

dat1<-read.table(text=" 
TIME, Value1, Value2 
01.08.2011 02:30:00, 4.4, 4.7 
01.09.2011 03:00:00, 4.2, 4.3 
01.11.2011 01:00:00, 3.5, 4.3 
01.12.2011 01:40:00, 3.4, 4.5 
01.01.2012 02:00:00, 4.8, 5.3 
01.02.2012 02:30:00, 4.9, 5.2 
01.08.2012 02:30:00, 4.1, 4.7 
01.12.2012 03:00:00, 4.7, 4.3 
01.01.2013 01:00:00, 3, 4.3 
01.01.2013 01:30:00, 3.8, 4.1 
01.01.2013 02:00:00, 3.8, 4.3 
01.01.2013 02:30:00, 3.9, 4.2 
01.01.2013 03:00:00, 3.7, 4.5 
01.01.2013 03:30:00, 3.5, 4.1 
01.02.2013 02:00:00, 3.8, 4.3
02.02.2013 04:30:00, 3.9, 4.2
",sep=",",header=TRUE,stringsAsFactors=FALSE)   
dat1$TIME<-as.POSIXct(dat1$TIME,format="%d.%m.%Y %H:%M:%S")

library(zoo) 
z1<- zoo(dat1[,-1],dat1[,1])

Vec<-format(seq(from=as.POSIXct("2012-01-01 02:30:00", format="%Y-%m-%d %H:%M:%S"),length.out=3, by="30 min"),"%H:%M:%S")
z1[format(time(z1),"%H:%M:%S")%in% Vec,]
#                    Value1 Value2
#2011-08-01 02:30:00    4.4    4.7
#2011-09-01 03:00:00    4.2    4.3
#2012-02-01 02:30:00    4.9    5.2
#2012-08-01 02:30:00    4.1    4.7
#2012-12-01 03:00:00    4.7    4.3
#2013-01-01 02:30:00    3.9    4.2
#2013-01-01 03:00:00    3.7    4.5
#2013-01-01 03:30:00    3.5    4.1
A.K.




----- Original Message -----
From: Jakob Hahn <Jakob.Hahn at stud.hs-esslingen.de>
To: arun <smartpink111 at yahoo.com>
Cc: 
Sent: Saturday, March 9, 2013 10:04 AM
Subject: Re: Zoo Data

Hi Arun,

z1[seq(which(time(z1)==as.POSIXct("**.**.2012 02:30:00",format="%d.%m.%Y %H:%M:%S")),which(time(z1)==as.POSIXct("**.**.2012 03:30:00",format="%d.%m.%Y %H:%M:%S"))),]

Is there a possibility to get all data e.g between two times - not depending on a special date - all values betweend 2:30 and 3:30?
Thanks
Jakob


---

Am 08.03.2013 um 22:05 schrieb arun:

> 
> 
> Hi,
> Try this:
>  z1[seq(which(time(z1)==as.POSIXct("01.01.2012 02:00:00",format="%d.%m.%Y %H:%M:%S")),which(time(z1)==as.POSIXct("01.01.2013 03:00:00",format="%d.%m.%Y %H:%M:%S"))),]
> 
> #                    Value1 Value2
> #2012-01-01 02:00:00    4.8    5.3
> #2012-02-01 02:30:00    4.9    5.2
> #2012-08-01 02:30:00    4.1    4.7
> #2012-12-01 03:00:00    4.7    4.3
> #2013-01-01 01:00:00    3.0    4.3
> #2013-01-01 01:30:00    3.8    4.1
> #2013-01-01 02:00:00    3.8    4.3
> #2013-01-01 02:30:00    3.9    4.2
> #2013-01-01 03:00:00    3.7    4.5
> A.K.
> 
> 
> ----- Original Message -----
> From: Jakob Hahn <Jakob.Hahn at gmail.com>
> To: arun <smartpink111 at yahoo.com>
> Cc: 
> Sent: Friday, March 8, 2013 3:40 PM
> Subject: Re: Zoo Data
> 
> Okay,
> but works only by line not by date - if I don't know which line my date is;-)
> Jakob
> 
> 
> ----
> Am 08.03.2013 um 21:32 schrieb arun:
> 
>> HI Jakob,
>> 
>> No problem.
>> 
>> Your statement
>> " Is it also possible plot from e.g. 01.01.2013 01:30:00 to 01.01.2013 02:00:00 - general spoken not the whole file?"
>> 
>> So, I subset the "whole dataset" to show that you can do the plot for a specific subset of your data. 
>> 
>> Depending upon the number of labels and its spacing, you can change it:
>> ix<- seq(1,length(tt), ....)
>> 
>> Arun
>> 
>> 
>> 
>> ----- Original Message -----
>> From: Jakob Hahn <Jakob.Hahn at gmail.com>
>> To: arun <smartpink111 at yahoo.com>
>> Cc: 
>> Sent: Friday, March 8, 2013 3:27 PM
>> Subject: Re: Zoo Data
>> 
>> Thats great, thank you very much!!!
>> Needed such a long time to find some stuff for this topic - I am a beginner but should work with R for my master thesis - so time is always against me;-)
>> Have a nice day, regards
>> Jakob
>> 
>> PS: What do you mean with the subset z2?
>> 
>> 
>> ---
>> Am 08.03.2013 um 21:11 schrieb arun:
>> 
>>> 
>>> 
>>> Hi Jakob,
>>> 
>>> dat1<-read.table(text=" 
>>> TIME, Value1, Value2
>>> 01.08.2011 02:30:00, 4.4, 4.7 
>>> 01.09.2011 03:00:00, 4.2, 4.3 
>>> 01.11.2011 01:00:00, 3.5, 4.3 
>>> 01.12.2011 01:40:00, 3.4, 4.5 
>>> 01.01.2012 02:00:00, 4.8, 5.3 
>>> 01.02.2012 02:30:00, 4.9, 5.2 
>>> 01.08.2012 02:30:00, 4.1, 4.7 
>>> 01.12.2012 03:00:00, 4.7, 4.3 
>>> 01.01.2013 01:00:00, 3, 4.3 
>>> 01.01.2013 01:30:00, 3.8, 4.1 
>>> 01.01.2013 02:00:00, 3.8, 4.3 
>>> 01.01.2013 02:30:00, 3.9, 4.2 
>>> 01.01.2013 03:00:00, 3.7, 4.5 
>>> 01.01.2013 03:30:00, 3.5, 4.1 
>>> 01.02.2013 02:00:00, 3.8, 4.3 
>>> ",sep=",",header=TRUE,stringsAsFactors=FALSE)  
>>> dat1$TIME<-as.POSIXct(dat1$TIME,format="%d.%m.%Y %H:%M:%S") 
>>> library(zoo) 
>>> z1<- zoo(dat1[,-1],dat1[,1])
>>>    plot(z1,plot.type="single",col=1:2,xaxt="n")
>>> tt<- time(z1)
>>> axis(side=1,at=tt,labels=FALSE)
>>>    ix<- seq(1,length(tt),4) #can change here
>>> fmt<- "%b-%Y"
>>> labs<- format(tt,fmt)
>>> axis(side=1,at=tt[ix],labels=labs[ix],tcl=-0.7,cex.axis=0.7)
>>> legend("topleft",colnames(z1),lty=1,col=1:2)
>>> 
>>> #subset
>>> z2<- zoo(dat1[1:5,-1],dat1[1:5,1])
>>> 
>>> Hope this helps.
>>> 
>>> A.K.
>>> 
>>> 
>>> 
>>> 
>>> ----- Original Message -----
>>> From: "jakob.hahn at gmail.com" <jakob.hahn at gmail.com>
>>> To: smartpink111 at yahoo.com
>>> Cc: 
>>> Sent: Friday, March 8, 2013 10:05 AM
>>> Subject: Zoo Data
>>> 
>>> Hey Arun,
>>> thank you so far, maybe you can tell me 2 more things:
>>> 1st-how to deal with the xlim - right now R just makes me 2 dates on x-axis with Dez31 and Feb28 - how to define the interval.
>>> 2nd-I am plotting multiple lines with col - is it possible to generate a legend with the name/header of the column from data?
>>> Would be nice, if you could help me!
>>> Thanks
>>> Jakob
>>>



More information about the R-help mailing list