[R] Help make this simpler – count business day

[Gabor Grothendieck]

On Tue, Nov 9, 2010 at 8:49 PM, Gabor Grothendieck
<ggrothendieck at gmail.com> wrote:
> On Tue, Nov 9, 2010 at 11:49 AM, cameron <raymond.fu at invesco.com> wrote:
>>
>> Help make this simpler – count business day
>>
>>
>> I am a beginner in R and this is my first post
>>
>> Want to count the day in month.  For example
>>
>>                 Day
>> 2010-09-01   1      Wed
>> 2010-09-02   2  Thurs
>> 2010-09-03   3  Friday
>> 2010-09-07   4  Tuesday
>> 2010-09-08   5  Wed
>> 2010-09-09   6  Thursday
>> 2010-09-10   7  Friday
>>
>> #-------------------------
>> library(tseries)
>>
>> msft <- get.hist.quote(instrument="MSFT", start="1986-03-31",
>> end="2008-09-10", quote=c("O","H","L","C","A","V"), provider="yahoo",
>> retclass="zoo")
>>
>> # tail(msft)
>> #            Open  High   Low Close AdjClose   Volume
>> #2008-09-03 27.00 27.18 26.84 26.90    25.73 57127700
>> #2008-09-04 26.74 26.89 26.35 26.35    25.21 66141900
>> #2008-09-05 26.03 26.22 25.63 25.65    24.54 82305200
>> #2008-09-08 26.21 26.33 25.67 26.12    24.99 62110800
>> #2008-09-09 26.20 26.60 26.05 26.10    24.97 85735700
>> #2008-09-10 26.52 26.86 26.25 26.44    25.29 75064900
>>
>>
>> countday<-function(z)
>> {
>>        z <- cbind(z,rep(0,times=nrow(z)))
>>
>>        rng <- range(time(z))
>>        StartDate <- rng[1]
>>        EndDate   <- rng[2]
>>
>>        starty <- as.numeric(format(StartDate, "%Y"))
>>        endy   <- as.numeric(format(EndDate, "%Y"))
>>
>>        year <- starty
>>
>>        for (year in starty:endy){
>>                for (month in 1:12){
>>                        rows <- which(as.numeric(format(time(z),"%m")) == month &
>> as.numeric(format(time(z),"%Y")) == year )
>>                        temp <- z[rows,]
>>                        n <- 1:nrow(temp)
>>                        z[rows,ncol(z)] <- n
>>                }
>>        }
>>        colnames(z) <- c(colnames(z[,1:(ncol(z)-1)]),"Day")
>>        return(z)
>> }
>>
>> msft <- countday(msft)
>>
>>>msft
>> #            Open  High   Low Close AdjClose   Volume        Day
>> #2008-09-03 27.00 27.18 26.84 26.90    25.73 57127700   2
>> #2008-09-04 26.74 26.89 26.35 26.35    25.21 66141900   3
>> #2008-09-05 26.03 26.22 25.63 25.65    24.54 82305200   4
>> #2008-09-08 26.21 26.33 25.67 26.12    24.99 62110800   5
>> #2008-09-09 26.20 26.60 26.05 26.10    24.97 85735700   6
>> #2008-09-10 26.52 26.86 26.25 26.44    25.29 75064900   7
>
> Try this:
>
> msft$Day <- ave(1:nrow(msft), as.yearmon(time(msft)), FUN = seq_along)

Note that as.yearmon is from the zoo package:

library(zoo)


-- 
Statistics & Software Consulting
GKX Group, GKX Associates Inc.
tel: 1-877-GKX-GROUP
email: ggrothendieck at gmail.com