Skip to content

plot cummulative sum from calendar time

4 messages · gallon li, Gavin Simpson, jim holtman +1 more

#
On Wed, 2007-12-19 at 14:59 +0800, gallon li wrote:
Is this what you want?

## some example data
dat <- cumsum(rnorm(100))
start.end <- as.Date(c("01/01/2007", "31/12/2007"), format = "%d/%m/%Y")
dates <- seq(start.end[1], start.end[2], length = 100)

## plot the data
plot(dat ~ dates, type = "l")

HTH

G
#
Here is an example of plotting the number of observations per month.
You can change this to any period you want.

# create some test data
x.d <- as.Date("2000-1-1") + runif(1000, 1, 500)
# create buckets of one month (or whatever period you want)
x.cut <- cut(x.d, breaks=seq(as.Date('2000-1-1'), max(x.d)+7, by="1 month"))
# plot the data
barplot(table(x.cut), main="Observations Per Month")
On Dec 19, 2007 1:59 AM, gallon li <gallon.li at gmail.com> wrote:

  
    
#
I am not sure I really understand what you want but
will this work?

tt<-c("03-Nov-1997","09-Oct-1991","27-Aug-1992","01-Jul-1994","19-Jan-1990",
"12-Nov-1993","08-Oct-1993","10-Nov-1982","08-Dec-1986","23-Dec-1987","02-Aug-1995",
"20-Oct-1998","29-Apr-1991","16-Mar-1994","20-May-1991","28-Dec-1987","14-Jul-1999",
"27-Nov-1998","09-Sep-1999","26-Aug-1999","20-Jun-1997","05-May-1995","26-Mar-1998",
"15-Aug-1994","24-Jun-1996","02-Oct-1996","12-Aug-1985","24-Jun-1992","09-Jan-1991",
"04-Mar-1988")
tt1 <-  as.Date(tt, "%d-%b-%Y"))
aa <- cumsum(1:length(tt))
plot(tt1,aa)
--- gallon li <gallon.li at gmail.com> wrote: