[R] converting monthly to weekly without changing the original values

[Dimitri Liakhovitski]

Hello!
I have a code for converting monthly values into weekly values:

monthly<-data.frame(month=c(20100301,20100401,20100501,20100601,20100301,20100401,20100501,20100601),monthly.value=c(100,NA,200,300,10,NA,20,30),market=c("Market
A","Market A","Market A","Market A","Market B","Market B","Market
B","Market B"))
monthly$month<-as.character(monthly$month)
monthly$month<-as.Date(monthly$month,"%Y%m%d")
(monthly)

library(zoo)
z <- read.zoo(monthly, split = "market")
all.dates <- seq(start(z), as.Date(as.yearmon(end(z)), frac = 1), by = "day") ##
mondays <- all.dates[weekdays(all.dates) == "Monday"] ##
weeks <- na.locf(z, xout = mondays, na.rm = FALSE) ##
do.call(rbind, by(weeks, as.yearmon(mondays), function(x)
zoo(x/nrow(x), rownames(x))))


I have 2 questions:
1. How can I make the code above leave NAs that were present in
"monthly" as NAs in all weeks of the corresponding month - rather than
impute them as it currently does?
2. In addition to 1, how can I ask zoo to actually keep the original
(monthly) values in each week of that month - rather than divide the
monthly value by the number of weeks?


Thanks a lot!



-- 
Dimitri Liakhovitski
Ninah Consulting
www.ninah.com