Hello,
This is likely fairly silly question, and I apologize to whomever takes the
time to respond.
I am a relatively new user of R, on Windows XP, version 2.3.1.
Say I have a data table that looks like the following:
x
Date Location Amount Blue Green
1 01/01/2001 Central 1817 TRUE FALSE
2 01/02/2001 Central 20358 FALSE TRUE
3 05/08/2001 Central 16245 FALSE TRUE
4 02/02/2002 Western 112 TRUE FALSE
5 21/03/2002 Western 98756 TRUE FALSE
6 01/04/2002 Western 1598414 FALSE TRUE
7 07/01/2001 Western 1255 FALSE TRUE
8 20/10/2003 Central 16289 TRUE FALSE
9 21/10/2003 Eastern 10000 FALSE TRUE
10 22/10/2003 Eastern 98737 FALSE TRUE
11 23/10/2003 Eastern 198756 TRUE FALSE
12 24/10/2003 Eastern 98756 FALSE TRUE
13 25/10/2003 Eastern 65895 TRUE FALSE
14 26/10/2003 Eastern 2142266 FALSE TRUE
15 27/10/2003 North 98756 TRUE FALSE
16 28/10/2003 North 548236 FALSE TRUE
and I want to do some summaries by Fiscal year (or FY quarter).
Reading manuals and such, I cobbled this less than satisfactory bit together
to start to build a factor representing a fiscal year split:
y<-as.Date(x$Date,"%d/%m/%Y")
y<-as.matrix(y)
y$FY0203<-ifelse((y>=(as.Date("2002-03-21")))&(y<=(as.Date
("2003-04-01"))),"TRUE","FALSE")
the values seem correct, but aside from ugly, the data is not in the proper
format - with some more effort I might fix that... But my question is: is
there a more simple function available to select a range of dates? Or, at
least, a more elegant approach?
[[alternative HTML version deleted]]
Here are three alternative ways to get the fiscal year as a numeric value assuming: dd <- as.Date(x$Date,"%d/%m/%Y") # add one to year if month is past March as.numeric(format(dd, "%Y")) + (format(dd, "%m") > "03") # same but using POSIXlt # (Even though there are no time zones involved I have seen # situations where the time zone nevertheless crept in # where one would least expect it so even though I believe # this is correct I would nevertheless triple check # all your results if you use this one) with(as.POSIXlt(dd), 1900 + year + (mon > 2)) # this makes use of the zoo yearqtr class # which represents dates as year + 0, .25, .5 or .75 for # the four quarters library(zoo) as.numeric(ceiling(as.yearqtr(dd))) On 10/7/06, Ian Broom <broomian at gmail.com> wrote:> Hello, > > This is likely fairly silly question, and I apologize to whomever takes the > time to respond. > > I am a relatively new user of R, on Windows XP, version 2.3.1. > > Say I have a data table that looks like the following: > > x > Date Location Amount Blue Green > 1 01/01/2001 Central 1817 TRUE FALSE > 2 01/02/2001 Central 20358 FALSE TRUE > 3 05/08/2001 Central 16245 FALSE TRUE > 4 02/02/2002 Western 112 TRUE FALSE > 5 21/03/2002 Western 98756 TRUE FALSE > 6 01/04/2002 Western 1598414 FALSE TRUE > 7 07/01/2001 Western 1255 FALSE TRUE > 8 20/10/2003 Central 16289 TRUE FALSE > 9 21/10/2003 Eastern 10000 FALSE TRUE > 10 22/10/2003 Eastern 98737 FALSE TRUE > 11 23/10/2003 Eastern 198756 TRUE FALSE > 12 24/10/2003 Eastern 98756 FALSE TRUE > 13 25/10/2003 Eastern 65895 TRUE FALSE > 14 26/10/2003 Eastern 2142266 FALSE TRUE > 15 27/10/2003 North 98756 TRUE FALSE > 16 28/10/2003 North 548236 FALSE TRUE > > and I want to do some summaries by Fiscal year (or FY quarter). > Reading manuals and such, I cobbled this less than satisfactory bit together > to start to build a factor representing a fiscal year split: > > y<-as.Date(x$Date,"%d/%m/%Y") > y<-as.matrix(y) > y$FY0203<-ifelse((y>=(as.Date("2002-03-21")))&(y<=(as.Date > ("2003-04-01"))),"TRUE","FALSE") > > the values seem correct, but aside from ugly, the data is not in the proper > format - with some more effort I might fix that... But my question is: is > there a more simple function available to select a range of dates? Or, at > least, a more elegant approach? > > [[alternative HTML version deleted]] > > ______________________________________________ > R-help at stat.math.ethz.ch mailing list > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide http://www.R-project.org/posting-guide.html > and provide commented, minimal, self-contained, reproducible code. >
And here is a fourth: as.numeric(format(dd, "%Y")) + (quarters(dd) > "Q1") On 10/8/06, Gabor Grothendieck <ggrothendieck at gmail.com> wrote:> Here are three alternative ways to get the fiscal year as a numeric > value assuming: > dd <- as.Date(x$Date,"%d/%m/%Y") > > # add one to year if month is past March > as.numeric(format(dd, "%Y")) + (format(dd, "%m") > "03") > > # same but using POSIXlt > # (Even though there are no time zones involved I have seen > # situations where the time zone nevertheless crept in > # where one would least expect it so even though I believe > # this is correct I would nevertheless triple check > # all your results if you use this one) > with(as.POSIXlt(dd), 1900 + year + (mon > 2)) > > # this makes use of the zoo yearqtr class > # which represents dates as year + 0, .25, .5 or .75 for > # the four quarters > library(zoo) > as.numeric(ceiling(as.yearqtr(dd))) > > > > On 10/7/06, Ian Broom <broomian at gmail.com> wrote: > > Hello, > > > > This is likely fairly silly question, and I apologize to whomever takes the > > time to respond. > > > > I am a relatively new user of R, on Windows XP, version 2.3.1. > > > > Say I have a data table that looks like the following: > > > > x > > Date Location Amount Blue Green > > 1 01/01/2001 Central 1817 TRUE FALSE > > 2 01/02/2001 Central 20358 FALSE TRUE > > 3 05/08/2001 Central 16245 FALSE TRUE > > 4 02/02/2002 Western 112 TRUE FALSE > > 5 21/03/2002 Western 98756 TRUE FALSE > > 6 01/04/2002 Western 1598414 FALSE TRUE > > 7 07/01/2001 Western 1255 FALSE TRUE > > 8 20/10/2003 Central 16289 TRUE FALSE > > 9 21/10/2003 Eastern 10000 FALSE TRUE > > 10 22/10/2003 Eastern 98737 FALSE TRUE > > 11 23/10/2003 Eastern 198756 TRUE FALSE > > 12 24/10/2003 Eastern 98756 FALSE TRUE > > 13 25/10/2003 Eastern 65895 TRUE FALSE > > 14 26/10/2003 Eastern 2142266 FALSE TRUE > > 15 27/10/2003 North 98756 TRUE FALSE > > 16 28/10/2003 North 548236 FALSE TRUE > > > > and I want to do some summaries by Fiscal year (or FY quarter). > > Reading manuals and such, I cobbled this less than satisfactory bit together > > to start to build a factor representing a fiscal year split: > > > > y<-as.Date(x$Date,"%d/%m/%Y") > > y<-as.matrix(y) > > y$FY0203<-ifelse((y>=(as.Date("2002-03-21")))&(y<=(as.Date > > ("2003-04-01"))),"TRUE","FALSE") > > > > the values seem correct, but aside from ugly, the data is not in the proper > > format - with some more effort I might fix that... But my question is: is > > there a more simple function available to select a range of dates? Or, at > > least, a more elegant approach? > > > > [[alternative HTML version deleted]] > > > > ______________________________________________ > > R-help at 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. > > >