R help - Feb 2007 - Working with temporal data

Hi,

I have several files with data in this format:

20070102
20070102
20070106
20070201
...

The data is sorted and each line represents a date (YYYYMMDD). I would
like to analyze this data using R. For instance, I would like to have
a histogram by year, month or day.

I've already made a simple Perl script that aggregates this data but I
believe that R can be much more powerful and easy on this kind of
work.

Any suggestions on where to start?

Thanks in advance,
S?rgio Nunes

Here is a start on what you want to do.  This generates some test data and
then does a couple of summaries:

> # generate some data
> N <- 1000
> x <- data.frame(date=as.character(20070000 + sample(1:4, N, TRUE) * 100
+
sample(1:31, N, TRUE)),
+     value=runif(N))
> head(x)  # display the data
      date      value
1 20070124 0.07904540
2 20070117 0.17864565
3 20070109 0.86078870
4 20070205 0.93952259
5 20070112 0.87904425
6 20070323 0.01717623
> # assuming you read it in as character, convert to Date for processing
> x$date <- as.Date(strptime(x$date, "%Y%m%d"))
> x <- x[order(x$date), ] # order by date for plotting
> plot(x$date, x$value, type='l')   # plot the data
> # show counts by month
> table(months(x$date))
   April February  January    March
     238      236      253      237
> # average by month
> aggregate(x$value, list(months(x$date)), mean)
   Group.1         x
1    April 0.4791387
2 February 0.5010831
3  January 0.5114135
4    March 0.4695668
>


On 2/15/07, Sérgio Nunes <snunes@gmail.com> wrote:
>
> Hi,
>
> I have several files with data in this format:
>
> 20070102
> 20070102
> 20070106
> 20070201
> ...
>
> The data is sorted and each line represents a date (YYYYMMDD). I would
> like to analyze this data using R. For instance, I would like to have
> a histogram by year, month or day.
>
> I've already made a simple Perl script that aggregates this data but I
> believe that R can be much more powerful and easy on this kind of
> work.
>
> Any suggestions on where to start?
>
> Thanks in advance,
> Sérgio Nunes
>
> ______________________________________________
> R-help@stat.math.ethz.ch 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.
>


-- 
Jim Holtman
Cincinnati, OH
+1 513 646 9390

What is the problem you are trying to solve?

	[[alternative HTML version deleted]]

S?rgio Nunes napsal(a):
> Hi,
> 
> I have several files with data in this format:
> 
> 20070102
> 20070102
> 20070106
> 20070201
> ...
> 
> The data is sorted and each line represents a date (YYYYMMDD). I would
> like to analyze this data using R. For instance, I would like to have
> a histogram by year, month or day.
> 
> I've already made a simple Perl script that aggregates this data but I
> believe that R can be much more powerful and easy on this kind of
> work.
> 
> Any suggestions on where to start?
> 
> Thanks in advance,
> S?rgio Nunes
Analysing a particular day is easy - just call e.g. 
hist(your.variable[day==20070101])
if 'day' contains the date stored as an integer.

If you want to do this for all days probably you mmight use a loop 
through unique(day)

Finally, to do a monthly/yearly analysis, just create 'month' and
'year'
month <- trunc(day/100)
year <- trunc(day/10000)

HTH
Petr Klasterecky
-- 
Dept. of Probability and Statistics
Charles University in Prague
Czech Republic

Hi again,

I'm still trying to read my data but I'm having some difficulties
converting it to dates.
My data file has lines and in each line a single date exists in the
format >2007/02/16< (without the >,<). I've tried the following:
> d <- readLines("file.dat")
> d
[1] "2006/08/09" "2004/02/11" "2004/06/09" ...
[2] ...
> d2 <- as.Date(d, format="%Y/%m/$d")
> d2
[1] NA NA NA ...
...

I'm surely doing something wrong.
Any advice would be welcomed.

Thanks!
S?rgio Nunes

On 2/15/07, S?rgio Nunes <snunes at gmail.com>
wrote:
> Hi,
>
> I have several files with data in this format:
>
> 20070102
> 20070102
> 20070106
> 20070201
> ...
>
> The data is sorted and each line represents a date (YYYYMMDD). I would
> like to analyze this data using R. For instance, I would like to have
> a histogram by year, month or day.
>
> I've already made a simple Perl script that aggregates this data but I
> believe that R can be much more powerful and easy on this kind of
> work.
>
> Any suggestions on where to start?
>
> Thanks in advance,
> S?rgio Nunes
>