Fill NA values with the trailing row value times a growth rate?

Question

What would be a good way to populate NA values with the previous value times (1 + growth)?

df <- data.frame(
  year = 0:6,
  price1 = c(1.1, 2.1, 3.2, 4.8, NA, NA, NA),
  price2 = c(1.1, 2.1, 3.2, NA, NA, NA, NA)
)
growth <- .02

In this case, I would want the missing values in price1 to be filled with 4.8*1.02, 4.8*1.02^2, and 4.8*1.02^3. Similarly, I would want the missing values in price2 to be filled with 3.2*1.02, 3.2*1.02^2, 3.2*1.02^3, and 3.2*1.02^4.

I've tried this, but I think it needs to be set to repeat somehow (apply?):

library(dplyr)
df %>%
  mutate(price1 = ifelse(is.na(price1),
    lag(price1) * (1 + growth), price1
  ))

I'm not using dplyr for anything else (yet), so something from base R or plyr or similar would be appreciated.

Answer 1

Assuming only trailing NAs:

NAgrow <- function(x,growth=0.02) {
    isna <- is.na(x)
    lastval <- tail(x[!isna],1)
    x[isna] <- lastval*(1+growth)^seq(sum(isna))
    return(x)
}

If there are interior NA values as well this would get a little trickier.

Apply to all columns except the first:

df[-1] <- lapply(df[-1],NAgrow)

##   year   price1   price2
## 1    0 1.100000 1.100000
## 2    1 2.100000 2.100000
## 3    2 3.200000 3.200000
## 4    3 4.800000 3.264000
## 5    4 4.896000 3.329280
## 6    5 4.993920 3.395866
## 7    6 5.093798 3.463783

Answer 2

A compact base R solution can be obtained using Reduce:

growthfun <- function(x, y) if (is.na(y)) (1+growth)*x else y
replace(df, TRUE, lapply(df, Reduce, f = growthfun, acc = TRUE))

giving:

  year   price1   price2
1    0 1.100000 1.100000
2    1 2.100000 2.100000
3    2 3.200000 3.200000
4    3 4.800000 3.264000
5    4 4.896000 3.329280
6    5 4.993920 3.395866
7    6 5.093798 3.463783

Note: The data in the question has no non-trailing NA values but if there were some then we could use na.fill from zoo to first replace the trailing NAs with a special value, such as NaN, and look for it instead of NA:

library(zoo)

DF <- as.data.frame(na.fill(df, c(NA, NA, NaN)))
growthfun <- function(x, y) if (is.nan(y)) (1+growth)*x else y
replace(DF, TRUE, lapply(DF, Reduce, f = growthfun, acc = TRUE))

Answer 3

The following solution based on rle works with NA in any position and does not rely on looping to fill in the missing values:

NAgrow.rle <- function(x) {
  if (is.na(x[1]))  stop("Can't have NA at beginning")
  r <- rle(is.na(x))
  na.loc <- which(r$values)
  b <- rep(cumsum(r$lengths)[na.loc-1], r$lengths[na.loc])
  x[is.na(x)] <- ave(x[b], b, FUN=function(y) y[1]*(1+growth)^seq_along(y))
  x
}
df[,-1] <- lapply(df[,-1], NAgrow.rle)
#   year   price1   price2
# 1    0 1.100000 1.100000
# 2    1 2.100000 2.100000
# 3    2 3.200000 3.200000
# 4    3 4.800000 3.264000
# 5    4 4.896000 3.329280
# 6    5 4.993920 3.395866
# 7    6 5.093798 3.463783

I'll drop in two additional solutions using for loops, one in base R and one in Rcpp:

NAgrow.for <- function(x) {
  for (i in which(is.na(x))) {
    x[i] <- x[i-1] * (1+growth)
  }
  x
}

library(Rcpp)
cppFunction(
"NumericVector NAgrowRcpp(NumericVector x, double growth) {
  const int n = x.size();
  NumericVector y(x);
  for (int i=1; i < n; ++i) {
    if (R_IsNA(x[i])) {
      y[i] = (1.0 + growth) * y[i-1];
    }
  }
  return y;
}")

The solutions based on rle (crimson and josilber.rle) take about twice as long as the simple solution based on a for loop (josilber.for), and as expected the Rcpp solution is the fastest, running in about 0.002 seconds.

set.seed(144)
big.df <- data.frame(ID=1:100000,
                     price1=sample(c(1:10, NA), 100000, replace=TRUE),
                     price2=sample(c(1:10, NA), 100000, replace=TRUE))
crimson <- function(df) apply(df[,-1], 2, function(x){
  if(sum(is.na(x)) == 0){return(x)}
  ## updated with optimized portion from @josilber
  r <- rle(is.na(x))
  na.loc <- which(r$values)
  b <- rep(cumsum(r$lengths)[na.loc-1], r$lengths[na.loc])
  lastValIs <- 1:length(x)
  lastValIs[is.na(x)] <- b
  x[is.na(x)] <-
    sapply(which(is.na(x)), function(i){
      return(x[lastValIs[i]]*(1 + growth)^(i - lastValIs[i]))
    })
  return(x)
})
ggrothendieck <- function(df) {
  growthfun <- function(x, y) if (is.na(y)) (1+growth)*x else y
  lapply(df[,-1], Reduce, f = growthfun, acc = TRUE)
}
josilber.rle <- function(df) lapply(df[,-1], NAgrow.rle)
josilber.for <- function(df) lapply(df[,-1], NAgrow.for)
josilber.rcpp <- function(df) lapply(df[,-1], NAgrowRcpp, growth=growth)
library(microbenchmark)
microbenchmark(crimson(big.df), ggrothendieck(big.df), josilber.rle(big.df), josilber.for(big.df), josilber.rcpp(big.df))
# Unit: milliseconds
#                   expr        min         lq       mean     median         uq         max neval
#        crimson(big.df)  98.447546 131.063713 161.494366 152.477661 183.175840  379.643222   100
#  ggrothendieck(big.df) 437.015693 667.760401 822.530745 817.864707 925.974019 1607.352929   100
#   josilber.rle(big.df)  59.678527 115.220519 132.874030 127.476340 151.665657  262.003756   100
#   josilber.for(big.df)  21.076516  57.479169  73.860913  72.959536  84.846912  178.412591   100
#  josilber.rcpp(big.df)   1.248793   1.894723   2.373469   2.190545   2.697246    5.646878   100