calculate time from dates

Dear Rui, I think your solution is the best. however, to keep only the
last one if month diff are equal part: because my data also have other
variables besides ID and month, your code doesn't work. where should i
modify it? in another word, my "dat2" contains more than 2 columns.i
tried to modify [1:2] to [1:p] where p is the total number of columns in
dat2 but it still doesn't work.

Best, Gallon


On Thu, Jul 11, 2013 at 7:00 PM, Rui Barradas <ruipbarradas at sapo.pt
<mailto:ruipbarradas at sapo.pt>> wrote:

    Hello,

    The functions in stackoverflow need a date 'format' argument.



    # Functions from
    #
    http://stackoverflow.com/__questions/1995933/number-of-__months-between-two-dates
    <http://stackoverflow.com/questions/1995933/number-of-months-between-two-dates>
    # with a 'format' argument added
    #
    # turn a date into a 'monthnumber' relative to an origin
    monnb <- function(d, format = "%Y-%m-%d") {
             lt <- as.POSIXlt(as.Date(d, origin="1900-01-01", format =
    format))
             lt$year*12 + lt$mon
    }
    # compute a month difference as a difference between two monnb's
    mondf <- function(d1, d2, format = "%Y-%m-%d") {
             monnb(d2, format = format) - monnb(d1, format = format)
    }


    dat1 <- read.table(text = "

    ID date
    1 4/12/2008
    1 4/13/2008
    1 5/11/2008
    2 3/21/2009
    2 4/22/2009
    2 8/05/2009
    ", header = TRUE)

    dat2 <- data.frame(ID = dat1$ID, month = mondf("01/01/2008",
    dat1$date, format = "%m/%d/%Y") + 1)

    # Now keep just the last one if month diffs are equal
    result <- with(dat2, aggregate(month, list(ID, month), FUN = tail,
    1))[1:2]
    names(result) <- names(dat2)
    result


    Hope this helps,

    Rui Barradas

    Em 11-07-2013 11:03, andrija djurovic escreveu:

        Hi.

        See
        http://stackoverflow.com/__questions/1995933/number-of-__months-between-two-dates
        <http://stackoverflow.com/questions/1995933/number-of-months-between-two-dates>

        Andrija


        On Thu, Jul 11, 2013 at 11:56 AM, Gallon Li <gallon.li at gmail.com
        <mailto:gallon.li at gmail.com>> wrote:

            My data are from 2008 to 2010, with repeated measures for
            same subjects. I
            wish to compute number of months since january 2008.

            The data are like the following:

            ID date
            1 4/12/2008
            1 4/13/2008
            1 5/11/2008
            2 3/21/2009
            2 4/22/2009
            2 8/05/2009
            ...

            the date column are in the format "%m/%d/%y". i wish to obtain

            ID month
            1 4
            1 4
            1 5
            2 15
            2 16
            2 20
            ...

            also, for the same ID with two identical month, I only want
            to keep the
            last one. can any expert help with this question?

                      [[alternative HTML version deleted]]