CODEWITHSUNDEEP
rfinance
runningbirds
runningbirds

Reputation: 6615

cumulative percentage return in r

I have a data frame with a variety of stock returns over a period of time. The returns are in percent gain or loss (.02 for 2% return or 102% of the previous periods value).

I am looking for either a function or method to cumulatively show the returns for each period (in percentages). For example, this would show the cumulative/compounding gains for stock1 to be .02, .0404, .09242 for the first 3 periods.... (1.02*1.02*1.05). 

   mydf = data.frame(period = c('a','b','c','d','e','f'), stock1=c(.02, .02, .05,-.05,-.05,0), stock2=c(0, .01,0,.03,.05,.01))
   mydf
   #help mydf$stk1_percentgain =

Upvotes: 0

Views: 2913

Answers (2)

Steven Beaupré
Steven Beaupré

Reputation: 21621

You could use dplyr:

mydf %>% 
  mutate_each(funs(
    paste0(formatC(100 * (cumprod(1 + .) - 1), format = "f", 2), "%")), -period)

Which gives:

#  period stock1 stock2
#1      a  2.00%  0.00%
#2      b  4.04%  1.00%
#3      c  9.24%  1.00%
#4      d  3.78%  4.03%
#5      e -1.41%  9.23%
#6      f -1.41% 10.32%

Upvotes: 3

eipi10
eipi10

Reputation: 93811

This will give you the cumulative return by period:

sapply(mydf[,-1], function(x) cumprod(1 + x) - 1)

          stock1    stock2
[1,]  0.02000000 0.0000000
[2,]  0.04040000 0.0100000
[3,]  0.09242000 0.0100000
[4,]  0.03779900 0.0403000
[5,] -0.01409095 0.0923150
[6,] -0.01409095 0.1032382

Or if you want something that's more human-readable:

sapply(mydf[,-1], function(x) paste0(sprintf("%0.2f", (cumprod(1 + x) - 1)*100, 2),"%"))

     stock1   stock2  
[1,] "2.00%"  "0.00%" 
[2,] "4.04%"  "1.00%" 
[3,] "9.24%"  "1.00%" 
[4,] "3.78%"  "4.03%" 
[5,] "-1.41%" "9.23%" 
[6,] "-1.41%" "10.32%"

Upvotes: 5

Related Questions