Row by row computation - r

Question

I have started learning R.
Need your help on how to do this computation in loop:
C1 C2 C3
A 5 10
A 4 6
B 2 3
B 8 9
I would like the output as
C1 C2 C3 C4(c2*c3)
A 5 10 50
A 4+50=54 6 54*6= 324
B 2 3 6
B 8+6=14 9 14*9 = 126
For each distinct value in column C1 the value calculated at C4 should be added to the new row value at C2.

Answer 1

One dplyr possibility could be:
df %>%
group_by(C1) %>%
mutate(C4 = (C2 + lag(C2 * C3, default = 0)) * C3,
C2 = C2 + lag(C2 * C3, default = 0))
C1 C2 C3 C4
<chr> <dbl> <int> <dbl>
1 A 5 10 50
2 A 54 6 324
3 B 2 3 6
4 B 14 9 126
Or using data.table (by #markus):
setDT(df)[, `:=`(C4 = (C2 + shift(C2 * C3, fill = 0)) * C3,
C2 = C2 + shift(C4, fill = 0)), by = C1]

Answer 2

As Loop ;-):
d <- read.table(text = "C1 C2 C3
A 5 10
A 4 6
B 2 3
B 8 9", header = TRUE)
my_calc <- function(d) {
d[, 4] <- d[, 2] * d[, 3]
for (i in seq_len(nrow(d) - 1)) {
d[i+1, 2] <- d[i, 4] + d[i+1, 2]
d[i+1, 4] <- d[i+1, 2] * d[i+1, 3]
}
d
}
do.call(rbind, lapply(split(d, d$C1), my_calc))
#C1 C2 C3 V4
#A.1 A 5 10 50
#A.2 A 54 6 324
#B.3 B 2 3 6
#B.4 B 14 9 126

Answer 3

This loop should do the trick:
df = data.frame(C1=LETTERS[c(1,1,2,2)],C2=c(5,4,2,8),C3=c(10,6,3,9),stringsAsFactors = FALSE)
df$C4=rep(0,nrow(df))
for (i in 1:nrow(df)){
if (i!=1){
if (df$C1[i]==df$C1[i-1]){
df$C2[i]=df$C2[i]+df$C4[i-1]
}
}
df$C4[i] = df$C2[i]*df$C3[i]
}
C1 C2 C3 C4
1 A 5 10 50
2 A 54 6 324
3 B 2 3 6
4 B 14 9 126