how to compute rolling averages

[zeeshan0112]

df <- c(2,4,1,5,9,5,7,5,9,8,7,5,6,8,4,2,8,1,9,3,8,5,11,16,10)

I need to get the average of the df in ascending order with 20% observation each. means separate average of 5 observation in ascending or descending order.

Thanks in advance

[technocrat]

suppressPackageStartupMessages({
  library(zoo)
})

dat <- c(2,4,1,5,9,5,7,5,9,8,7,5,6,8,4,2,8,1,9,3,8,5,11,16,10)
rollmean(dat,5)
#>  [1]  4.2  4.8  5.4  6.2  7.0  6.8  7.2  6.8  7.0  6.8  6.0  5.0  5.6  4.6  4.8
#> [16]  4.6  5.8  5.2  7.2  8.6 10.0

[Yarnabrina]

Another option that I learned today from SO:

> dat <- c(2, 4, 1, 5, 9, 5, 7, 5, 9, 8, 7, 5, 6, 8, 4, 2, 8, 1, 9, 3, 8, 5, 11, 16, 10)
> as.numeric(x = stats::filter(x = dat, rep(x = (1 / 5), each = 5), sides = 2))
 [1]   NA   NA  4.2  4.8  5.4  6.2  7.0  6.8  7.2  6.8  7.0  6.8  6.0  5.0  5.6  4.6  4.8
[18]  4.6  5.8  5.2  7.2  8.6 10.0   NA   NA

[davis]

If you've used the purrr package and enjoy its API, you might like slider

library(slider)
library(tibble)

df <- tibble(
  x = c(2,4,1,5,9,5,7,5,9,8,7,5,6,8,4,2,8,1,9,3,8,5,11,16,10),
  # Current value + 4 before it
  mean = slide_mean(x, before = 4, complete = TRUE)
)

df
#> # A tibble: 25 x 2
#>        x  mean
#>    <dbl> <dbl>
#>  1     2  NA  
#>  2     4  NA  
#>  3     1  NA  
#>  4     5  NA  
#>  5     9   4.2
#>  6     5   4.8
#>  7     7   5.4
#>  8     5   6.2
#>  9     9   7  
#> 10     8   6.8
#> # … with 15 more rows

^{Created on 2021-05-06 by the reprex package (v1.0.0)}