I have weekly time series data with year, month, day, and price variables. The input data set for the weekly series takes the following form: Year month day price 1990 8 20 119.1 1990 8 27 124.5 1990 9 3 124.2 1990 9 10 125.2 1990 9 17 126.6 1990 9 24 127.2 1990 10 1 132.1 1990 10 8 133.3 1990 10 15 133.9 1990 10 22 134.5 1990 10 29 133.9 .. ... ... ... ... ... .... .... 2008 3 3 313.7 2008 3 10 320 2008 3 17 325.7 2008 3 24 322.4 I would like to collapse the data into monthly averages to merge with a monthly series. The input data set for the monthly series takes the following form: M-Y Year Month Change Aug-1990 1990 8 -226.871 Sep-1990 1990 9 -896.333 Oct-1990 1990 10 111.419 Nov-1990 1990 11 -364.2 Dec-1990 1990 12 -527.645 Jan-1991 1991 1 -70.935 Feb-1991 1991 2 231.214 Mar-1991 1991 3 -239 ... ... ..... .. .... The merged data set should be of class(ts). I can perform the conversions outside of R and then import but I would rather perform all conversions within R. I have looked through the zoo and Rmetrics packages but without success. Any help will be appreciated. Thanks, Richard Saba Department of Economics Auburn University Email: sabaric at auburn.edu Phone: 334 844-2922
If I understand:
x.new <- cbind('M.Y'=paste(month.abb[x$month], x$Year,
sep="-"), x[-3])
aggregate(list(Change=x.new$price),
list("M.Y"=x.new[,"M.Y"], Year=x.new$Year,
Month=x.new$month), FUN=mean)
On 30/03/2008, Richard Saba <sabaric at auburn.edu>
wrote:>
> I have weekly time series data with year, month, day, and price variables.
> The input data set for the weekly series takes the following form:
>
> Year month day price
> 1990 8 20 119.1
> 1990 8 27 124.5
> 1990 9 3 124.2
> 1990 9 10 125.2
> 1990 9 17 126.6
> 1990 9 24 127.2
> 1990 10 1 132.1
> 1990 10 8 133.3
> 1990 10 15 133.9
> 1990 10 22 134.5
> 1990 10 29 133.9
> .. ... ... ...
> ... ... .... ....
> 2008 3 3 313.7
> 2008 3 10 320
> 2008 3 17 325.7
> 2008 3 24 322.4
>
>
> I would like to collapse the data into monthly averages to merge with a
> monthly series. The input data set for the monthly series takes the
> following form:
>
> M-Y Year Month Change
> Aug-1990 1990 8 -226.871
> Sep-1990 1990 9 -896.333
> Oct-1990 1990 10 111.419
> Nov-1990 1990 11 -364.2
> Dec-1990 1990 12 -527.645
> Jan-1991 1991 1 -70.935
> Feb-1991 1991 2 231.214
> Mar-1991 1991 3 -239
>
> ... ... ..... .. ....
>
>
> The merged data set should be of class(ts).
> I can perform the conversions outside of R and then import but I would
> rather perform all conversions within R. I have looked through the zoo and
> Rmetrics packages but without success.
>
> Any help will be appreciated.
> Thanks,
>
> Richard Saba
> Department of Economics
> Auburn University
> Email: sabaric at auburn.edu
> Phone: 334 844-2922
>
> ______________________________________________
> 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.
>
--
Henrique Dallazuanna
Curitiba-Paran?-Brasil
25? 25' 40" S 49? 16' 22" O
Do you mean merge them into a two column series of price
for one column and change for another column? That is what
I will assume since the result was not illustrated.
Suggest you read the help files and the three
vignettes in the zoo package.
Also R News 4/1 has info on dates.
If you were looking for OHLC representations then see
the xts package which in turn uses zoo and read its
documentation in addition to the above.
library(zoo) # ensure you are using zoo 1.5.0 or later
price.Lines <- "Year month day price
1990 8 20 119.1
1990 8 27 124.5
1990 9 3 124.2
1990 9 10 125.2
1990 9 17 126.6
1990 9 24 127.2
1990 10 1 132.1
1990 10 8 133.3
1990 10 15 133.9
1990 10 22 134.5
1990 10 29 133.9
2008 3 3 313.7
2008 3 10 320
2008 3 17 325.7
2008 3 24 322.4
"
price.DF <- read.table(textConnection(price.Lines), header = TRUE)
price <- with(price.DF,
zoo(price, as.Date(paste(Year, month, day, sep = "-"))))
change.Lines <- "M-Y Year Month Change
Aug-1990 1990 8 -226.871
Sep-1990 1990 9 -896.333
Oct-1990 1990 10 111.419
Nov-1990 1990 11 -364.2
Dec-1990 1990 12 -527.645
Jan-1991 1991 1 -70.935
Feb-1991 1991 2 231.214
Mar-1991 1991 3 -239
"
change <- read.zoo(textConnection(change.Lines), header = TRUE,
FUN = as.yearmon, format = "%b-%Y",
colClasses = c("character", "NULL", "NULL",
"numeric"))
# convert change series to Date using last of month as the date
time(change) <- as.Date(time(change), frac = 1)
merge(price, change)
On Sun, Mar 30, 2008 at 11:24 AM, Richard Saba <sabaric at auburn.edu>
wrote:>
> I have weekly time series data with year, month, day, and price variables.
> The input data set for the weekly series takes the following form:
>
> Year month day price
> 1990 8 20 119.1
> 1990 8 27 124.5
> 1990 9 3 124.2
> 1990 9 10 125.2
> 1990 9 17 126.6
> 1990 9 24 127.2
> 1990 10 1 132.1
> 1990 10 8 133.3
> 1990 10 15 133.9
> 1990 10 22 134.5
> 1990 10 29 133.9
> .. ... ... ...
> ... ... .... ....
> 2008 3 3 313.7
> 2008 3 10 320
> 2008 3 17 325.7
> 2008 3 24 322.4
>
>
> I would like to collapse the data into monthly averages to merge with a
> monthly series. The input data set for the monthly series takes the
> following form:
>
> M-Y Year Month Change
> Aug-1990 1990 8 -226.871
> Sep-1990 1990 9 -896.333
> Oct-1990 1990 10 111.419
> Nov-1990 1990 11 -364.2
> Dec-1990 1990 12 -527.645
> Jan-1991 1991 1 -70.935
> Feb-1991 1991 2 231.214
> Mar-1991 1991 3 -239
>
> ... ... ..... .. ....
>
>
> The merged data set should be of class(ts).
> I can perform the conversions outside of R and then import but I would
> rather perform all conversions within R. I have looked through the zoo and
> Rmetrics packages but without success.
>
> Any help will be appreciated.
> Thanks,
>
> Richard Saba
> Department of Economics
> Auburn University
> Email: sabaric at auburn.edu
> Phone: 334 844-2922
>
> ______________________________________________
> 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.
>