R dplyr rolling sum

Question

i am implementing a rolling sum calculation through dplyr, but in my database i have a number of variables that have only one or only a few observations, causing an (k is sm

Answer 1

 library(dplyr)
 dg %>% 
   arrange(site,year,animal) %>% 
   group_by(site,animal) %>% 
   mutate(rollsum=cumsum(count))