I have a vector that looks something like this:
c(0.5,0,0,0,0,0.7,0,0,0,0,0.4,0,0,0,0)
Suppose I want to copy the values on positions 1, 6
Here's another base R approach. Initial zeros are left as is:
v = c(0,1,2,-2.1,0,3,0,0.4,0,0)
v[v!=0] = diff(c(0, v[v!=0]))
cumsum(v)
# [1] 0.0 1.0 2.0 -2.1 -2.1 3.0 3.0 0.4 0.4 0.4
And here are some benchmarks:
roland = function(v) {v[v == 0] <- NA; na.locf(v)}
mp = function(x) {with(rle(x), rep(replace(values, values==0, values[which(values == 0)-1]), lengths))}
quant = function(dat) {not.0 <- (dat != 0); approx(which(not.0), dat[not.0], xout = seq(along.with = dat), method = "constant", rule = 2)}
eddi = function(v) {v[v!=0] = diff(c(0, v[v!=0])); cumsum(v)}
v = sample(c(-10:10, 0), 1e6, TRUE)
microbenchmark(roland(v), mp(v), quant(v), eddi(v), times = 10)
#Unit: milliseconds
# expr min lq median uq max neval
# roland(v) 595.1630 625.7692 638.4395 650.4758 664.9224 10
# mp(v) 410.8224 433.6775 469.9346 496.6328 528.3218 10
# quant(v) 646.1775 753.0684 759.9805 838.4281 883.3383 10
# eddi(v) 265.8064 286.2922 316.7022 339.0333 354.0836 10