Hello:
1. I tried the example. The first thing I noticed was that
"library(fSeries)" loaded with the following comment:
# The new version of 'fSeries' has been renamed to 'timeSeries'
2. With this, I tried "library(timeSeries)". Then your first
example worked without error.
3. This time, "head(ts)" generated an error. To diagnose it, I
tried "str(ts)". This generated a summary that I found beautifully
formated but unintelligible. To understand the object, I then tried
"utils:::str.default(ts)". This described the object in terms I could
understand.
4. Another "heads up": "ts" is the name of a function in the
"stats" package. I would avoid creating an object by that name.
Hope this helps.
Spencer
David Winsemius wrote:
On Jul 12, 2009, at 1:05 PM, David Winsemius wrote:
On Jul 12, 2009, at 8:10 AM, tradenet wrote:
I added a reproducible example to my question...
ts<-dummyDailySeries(x = rnorm(365), units = NULL, zone = "",
FinCenter =
"")
"Heads up" is an English colloquial warning.
It may be reproducible on a machine that has all of your particular
loaded packages, but certainly not on one that doesn't.
ts<-dummyDailySeries(x = rnorm(365), units = NULL, zone = "",
FinCenter =
+ "")
Error: could not find function "dummyDailySeries"
Many people who read the list just ignore postings where the
questioner fails to indicate the package from which functions arise.
In penance for my failure to trim the r-help list address, I looked up
the function which is part of the fSeries package. I now wonder if
tradenet is using an outdated version. I get a different error after
installing and loading Package fSeries (270.76.1).
> ts<-dummyDailySeries(x = rnorm(365), units = NULL, zone = "",
FinCenter =
+ "")
Error in rulesFinCenter(FinCenter) :
There are no DST rules for GMT FinCenter!
Eventually I convince the function into returning a result and then
ts[1,0] fails with this error message:
> ts<-dummyDailySeries(x = rnorm(365), units = NULL, zone = "GMT",
FinCenter = "Zurich")
> head(ts)
TS.1
1970-01-01 01:00:00 -0.2463952
1970-01-02 01:00:00 0.9421668
1970-01-03 01:00:00 0.3354345
1970-01-04 01:00:00 1.7449670
1970-01-05 01:00:00 0.1551346
1970-01-06 01:00:00 -1.0082531
> ts[1,0]
Error in checkSlotAssignment(object, name, value) :
assignment of an object of class "NULL" is not valid for slot
"units" in an object of class "timeSeries"; is(value, "character") is
not TRUE
So now I am wondering if the package was recently redesigned to use S4
methods. My suggestion is to repost if you are still puzzled after
updating your copy of fSeriesand reviewing the timeSeries extractor
functions for S4 objects including seriesPositions():
X = timeSeries(matrix(rnorm(24), 12), timeCalendar())
as.character(seriesPositions(X)@Data[1])
[1] "2009-01-01" # a character vector ... which is what was requested
David Winsemius, MD
Heritage Laboratories
West Hartford, CT
______________________________________________
R-help@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.
______________________________________________
R-help@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.
