CODEWITHSUNDEEP
r
Vivi
Vivi

Reputation: 4218

How can I read a date series of quarterly data into R?

I have a dataset, the first column of which is the date in the format Year-Quarter, like the following

 1959-I   
 1959-II   
 1959-III   
 1959-IV    
 1960-I   
 1960-II   
 1960-III   
 1960-IV

I have imported the dataset into R, but I don't know how to convert this to Date format using the as.Date function, since there doesn't appear to be a conversion specification for quarters and I have just started learning R.

The best I can think of is something like

#extract the year
> dates <- substring(data$X,1,4)
> dates[1:8]
[1] "1959" "1959" "1959" "1959" "1960" "1960" "1960" "1960"
> dates <- as.numeric(dates)

#extract the quarter
> quarters <- substring(data$X,6)
> quarters[1:10]
[1] "I "   "II "  "III " "IV "  "I "   "II "  "III " "IV "

but this cannot possibly be the best way of doing this, and it still leaves me the problem of how to deal with my series quarters. To make matters worse, there is a blank character at the end of each character I don't know how to deal with because there are 3 different "lenghts" for the quarters strings.

Another option, of course, would be to fabricate my own quarters series, with something like

quarters <- rep(c(1,2,3,4),dates[length(dates)]-dates[1])

(I am quite proud of myself for having written this last line!).

So my question: is there a built-in way to import quarterly data in the format I have in R, and if not, any other format (still for quarterly data)? Any suggestions on the best way to proceed?

Upvotes: 10

Views: 21950

Answers (1)

G. Grothendieck
G. Grothendieck

Reputation: 269644

Suppose we have data 1:8 which corresponds to the 8 quarters you mention. Then we can do this:

ts(1:8, start = c(1959, 1), frequency = 4)
##      Qtr1 Qtr2 Qtr3 Qtr4
## 1959    1    2    3    4
## 1960    5    6    7    8

Also the zoo package has the "yearqtr" class:

 library(zoo)

 z <- zooreg(1:8, start = as.yearqtr("1959-1"), frequency = 4)
 z
 ## 1959 Q1 1959 Q2 1959 Q3 1959 Q4 1960 Q1 1960 Q2 1960 Q3 1960 Q4 
 ##       1       2       3       4       5       6       7       8

If we really did have a vector of such quarters that are not necessarily consecutive:

dt <- c("1959-I", "1959-II", "1959-III", "1959-IV", "1960-I", "1960-II", 
      "1960-III", "1960-IV")

we could convert them using gsubfn:

library(gsubfn)

g <- gsubfn("I.*", list(`I` = 1, `II` = 2, `III` = 3, `IV` = 4), dt)
g
## [1] "1959-1" "1959-2" "1959-3" "1959-4" "1960-1" "1960-2" "1960-3" "1960-4"

as.yearqtr(g)
## [1] "1959 Q1" "1959 Q2" "1959 Q3" "1959 Q4" "1960 Q1" "1960 Q2" "1960 Q3"
## [8] "1960 Q4"

Upvotes: 19

Related Questions