R help - Jun 2010 - setting up zoo objects from a data frame

Dear R People:

I have the following data frame:
> str(dog.df)
'data.frame':   7 obs. of  2 variables:
 $ V1: Factor w/ 3 levels "1/1/2000","1/2/2000",..: 1 1 1 2
2 3 3
 $ V2: Factor w/ 3 levels "cat","dog","tree": 2 1
3 1 3 2 3
> dog.df
        V1   V2
1 1/1/2000  dog
2 1/1/2000  cat
3 1/1/2000 tree
4 1/2/2000  cat
5 1/2/2000 tree
6 1/3/2000  dog
7 1/3/2000 tree
>
I would like to set up 3 time series; one for dog, one for cat, one
for tree, such that each runs from 1/1/2000 to 1/3/2000, with 0 if
there is no entry for the day.

So for instance, I would have:

dog.zoo
1/1/2000 1
1/2/2000 0
1/3/2000 1

Any help would be much appreciated.

Thanks in advance,
Sincerely,
Erin


-- 
Erin Hodgess
Associate Professor
Department of Computer and Mathematical Sciences
University of Houston - Downtown
mailto: erinm.hodgess at gmail.com

Am 08.06.2010 16:52, schrieb Erin Hodgess:
>
> I would like to set up 3 time series; one for dog, one for cat, one
> for tree, such that each runs from 1/1/2000 to 1/3/2000, with 0 if
> there is no entry for the day.
>
>
>   
Before using zoo or zooreg you should transform your data.frame as such
as that you have one column for each time series you want to have
afterwards. have a look at
?reshape
to do so.
little example:
test<-data.frame(date=c("2001/01/01","2001/01/01","2001/01/02","2001/01/02"),animal=c("dog","cat","dog","cat"),data=1:4)
test.w<-reshape(test,idvar="date",timevar="animal",direction="wide")
test.w

then you can do the zoo stuff, e.g. with
?zooreg
if you have a "regular" series.
It will transform automatically into a multiple time series.

hth
Stefan

The first time I posted this it got held up for approval so I am
trying it again.  Hopefully this time it gets through right away.

As with your prior post we can use read.zoo(..., split=...).
Alternatives are reshape and reshape::cast.


# read data into DF

Lines <- "V1   V2
1 1/1/2000  dog
2 1/1/2000  cat
3 1/1/2000 tree
4 1/2/2000  cat
5 1/2/2000 tree
6 1/3/2000  dog
7 1/3/2000 tree"

DF <- read.table(textConnection(Lines), header = TRUE)
DF$V1 <- as.Date(DF$V1, "%m/%d/%Y")

# 1. using zo

library(zoo)
d <- read.zoo(cbind(DF, 1), split = 2)
do.call(merge, c(d, fill = 0))

# 2. using reshape

DF2 <- with(DF, data.frame(Date = V1, V2, 1))
r <- reshape(DF2, dir = "wide", idvar = "Date", timevar =
"V2")
colnames(r) <- c("Date", sub("X1.", "",
colnames(r))[-1])
r[is.na(r)] <- 0
r

#  3. using the reshape package (where DF2 is as in #2)

library(reshape)
cast(formula = Date ~ V2, data = DF2, value = "X1", fill = 0)