[R] Import time series data with uneven dates

[Heywood, Giles]

A solution is at hand using the 'irts' (irreglar time-series) 
class from package tseries.

If your raw data is in a csv file, you could proceed as follows:

mydata <- read.csv(filename,header=TRUE)
basedate <- as.POSIXct(strptime("2003-01-01 00:00:00",format="%Y-%m-%d %X"))
rawdates <- basedate+(mydata[,1]-floor(mydata[,1]))*24*60*60*365 -60*60
truncdates <-
as.POSIXct(strptime(format.POSIXct(rawdates,format="%Y-%m-%d"),,format="%Y-%
m-%d"))
require(tseries)
x <- irts(time=truncdates,mydata[,2])

The result is an object of class 'irts'.

The format of the dates in your file is particularly unhelpful! If
they are available in a format recognisable by strptime, the above
could be simplified and made more robust.

Incidentally, I plan to post a slightly different 'its' (irregular 
time-series) package sometime soon.

Giles

> -----Original Message-----
> From: Ariel Andres [mailto:aandres1 at cogeco.ca]
> Sent: 19 June 2003 03:51
> To: r-help at stat.math.ethz.ch
> Subject: [R] Import time series data with uneven dates
> 
> 
> I am trying to import a file of daily index closing prices in 
> business time
> which excludes weekends and holidays so deltat is not 
> constant. My file
> looks like the following:
> date        close
> 2003.0055   47.05
> 2003.0082	45.71
> 2003.0164   43.45
> 2003.0192   42.96
> 2003.0219   44.56
> 2003.0247   42.99
> 2003.0274   42.28
> 2003.0356   41.74
> etc.
> 
> >From what I saw in the EuStockMarkets file, it appears that 
> they are also
> using years and fractions of years as the measure of time. 
> How do I read my
> file and still keep 'date' as the date?
> 
> Thanks!
> 
> A. Andres
> 95 Wellington St W
> Toronto, Ontario
> M5J 2N7
> 
> ______________________________________________
> R-help at stat.math.ethz.ch mailing list
> https://www.stat.math.ethz.ch/mailman/listinfo/r-help
> 


********************************************************************** 
This is a commercial communication from Commerzbank AG.\ \ This ... {{dropped}}