Hi everyone, I have a data frame, with the following format: MatDate-> row.names ID1 ID2 ID3 1 date1 2 date1 3 date3 etc but I cannot perform a rollapply() statement on the matrix without converting the matrix into a time series. i.e. MatTs<-ts(MatDate) Only then will my rollapply statement work: MatMin<-rollapply(MatTs, 2,by=2, min, na.rm=F) If I apply the rollapply() statement to the dataframe, I get the following error: Error: could not find function "rollapply" The problem is that when I convert the data.frame matrix into a time series matrix, I lose the dates in the row.names column. I just want to know if anyone could suggest a way to get around this problem, i.e. keep the row.names column in place, and use the rollapply() statement as above. Thanks, rcoder -- View this message in context: http://www.nabble.com/losing-row.names-in-matrix-operations-tp18788509p18788509.html Sent from the R help mailing list archive at Nabble.com.
If its regular you can convert it to ts or zoo. If its irregular convert it to zoo. There is no reason to expect rollapply to work with objects of other classes. Read ?ts and ?zoo. In ts note the start and frequency arguments. On Sat, Aug 2, 2008 at 7:50 AM, rcoder <mpdotbook at gmail.com> wrote:> > Hi everyone, > > I have a data frame, with the following format: > > MatDate-> > row.names ID1 ID2 ID3 > 1 date1 > 2 date1 > 3 date3 > etc > > but I cannot perform a rollapply() statement on the matrix without > converting the matrix into a time series. > i.e. MatTs<-ts(MatDate)Use the start and frequency arguments. See ?ts> > Only then will my rollapply statement work: > MatMin<-rollapply(MatTs, 2,by=2, min, na.rm=F) > > If I apply the rollapply() statement to the dataframe, I get the following > error: Error: could not find function "rollapply" > > The problem is that when I convert the data.frame matrix into a time series > matrix, I lose the dates in the row.names column. I just want to know if > anyone could suggest a way to get around this problem, i.e. keep the > row.names column in place, and use the rollapply() statement as above. > > Thanks, > > rcoder > -- > View this message in context: http://www.nabble.com/losing-row.names-in-matrix-operations-tp18788509p18788509.html > Sent from the R help mailing list archive at Nabble.com. > > ______________________________________________ > R-help at 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. >
Thanks for your reply Gabor. As I already have a dates column in my dataframe (in column row.names), is it possible to preserve this whilst still making a data set suitable for rollapply()? Thanks, rcoder Gabor Grothendieck wrote:> > If its regular you can convert it to ts or zoo. > If its irregular convert it to zoo. There is no > reason to expect rollapply to work with objects > of other classes. Read ?ts and ?zoo. In > ts note the start and frequency arguments. > > > On Sat, Aug 2, 2008 at 7:50 AM, rcoder <mpdotbook at gmail.com> wrote: >> >> Hi everyone, >> >> I have a data frame, with the following format: >> >> MatDate-> >> row.names ID1 ID2 ID3 >> 1 date1 >> 2 date1 >> 3 date3 >> etc >> >> but I cannot perform a rollapply() statement on the matrix without >> converting the matrix into a time series. >> i.e. MatTs<-ts(MatDate) > > Use the start and frequency arguments. See ?ts > >> >> Only then will my rollapply statement work: >> MatMin<-rollapply(MatTs, 2,by=2, min, na.rm=F) >> >> If I apply the rollapply() statement to the dataframe, I get the >> following >> error: Error: could not find function "rollapply" >> >> The problem is that when I convert the data.frame matrix into a time >> series >> matrix, I lose the dates in the row.names column. I just want to know if >> anyone could suggest a way to get around this problem, i.e. keep the >> row.names column in place, and use the rollapply() statement as above. >> >> Thanks, >> >> rcoder >> -- >> View this message in context: >> http://www.nabble.com/losing-row.names-in-matrix-operations-tp18788509p18788509.html >> Sent from the R help mailing list archive at Nabble.com. >> >> ______________________________________________ >> R-help at 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 at 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. > >-- View this message in context: http://www.nabble.com/losing-row.names-in-matrix-operations-tp18788509p18789688.html Sent from the R help mailing list archive at Nabble.com.
Constructing a zoo object is well covered in ?zoo
and in the three vignettes:
vignette("zoo")
vignette("zoo-quickref")
vignette("zoo-faq")
On Sat, Aug 2, 2008 at 10:27 AM, rcoder <mpdotbook at gmail.com>
wrote:>
> Thanks for your reply Gabor.
>
> As I already have a dates column in my dataframe (in column row.names), is
> it possible to preserve this whilst still making a data set suitable for
> rollapply()?
>
> Thanks,
>
> rcoder
>
>
>
> Gabor Grothendieck wrote:
>>
>> If its regular you can convert it to ts or zoo.
>> If its irregular convert it to zoo. There is no
>> reason to expect rollapply to work with objects
>> of other classes. Read ?ts and ?zoo. In
>> ts note the start and frequency arguments.
>>
>>
>> On Sat, Aug 2, 2008 at 7:50 AM, rcoder <mpdotbook at gmail.com>
wrote:
>>>
>>> Hi everyone,
>>>
>>> I have a data frame, with the following format:
>>>
>>> MatDate->
>>> row.names ID1 ID2 ID3
>>> 1 date1
>>> 2 date1
>>> 3 date3
>>> etc
>>>
>>> but I cannot perform a rollapply() statement on the matrix without
>>> converting the matrix into a time series.
>>> i.e. MatTs<-ts(MatDate)
>>
>> Use the start and frequency arguments. See ?ts
>>
>>>
>>> Only then will my rollapply statement work:
>>> MatMin<-rollapply(MatTs, 2,by=2, min, na.rm=F)
>>>
>>> If I apply the rollapply() statement to the dataframe, I get the
>>> following
>>> error: Error: could not find function "rollapply"
>>>
>>> The problem is that when I convert the data.frame matrix into a
time
>>> series
>>> matrix, I lose the dates in the row.names column. I just want to
know if
>>> anyone could suggest a way to get around this problem, i.e. keep
the
>>> row.names column in place, and use the rollapply() statement as
above.
>>>
>>> Thanks,
>>>
>>> rcoder
>>> --
>>> View this message in context:
>>>
http://www.nabble.com/losing-row.names-in-matrix-operations-tp18788509p18788509.html
>>> Sent from the R help mailing list archive at Nabble.com.
>>>
>>> ______________________________________________
>>> R-help at 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 at 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.
>>
>>
>
> --
> View this message in context:
http://www.nabble.com/losing-row.names-in-matrix-operations-tp18788509p18789688.html
> Sent from the R help mailing list archive at Nabble.com.
>
> ______________________________________________
> R-help at 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.
>