Using runner::runner to extract sliding windows from timeseries

I want to calculate a parameter (not with a standard function like mean()) for sliding windows. I have found the runner::runner() function, which is probably capable of providing the functionality that I want, i.e. sliding windows based on time, robust in case of missing dates and uneven spacing, but I cannot get the right syntax.

Please see my reprex below, where I try to get windows of three days in length with a lag of 1 day. So window_nr1 would be: d0 - d3, window_nr2: d1 - d4 etc.:

Data:
library(runner)
t <- seq(from=0, to=(3600*24*5), by=3600)
runner(x=t, k=(3600*24*3), lag=(3600*24), idx=t)

Expected output:

[[1]]
[1] 0 ...... [72] 259200
...
[[3]]
[1] 172800 .... [72] 432000
...
## --> one window spans three days, with one day overlap

Instead I get this:

[[96]]
[1] 0 ..... [72] 255600
[[97]]
[1] 3600 ... [72] 259200
[[98]]
[1] 7200 .... [72] 262800
...
## --> basically no matter what I enter for lag, 
##     the window shifts one hour (3600s), or one step along `t`.

I have one major and one minor question:

• how to I get runner to shift the window one day(86400 s), and not one step along the t-axis?
• how to discard the first 95 elements of the result-set that are filled with windows who do not include full 3 days?

You need at instead of lag; we can also use Filter() to filter out elements that are less than 3 days long.

To make this easier to comprehend, I used hours instead of seconds, but you can simply multiply the output by 3600.

library(runner)
t <- seq(from=0, to=(24*5), by=1)

Filter(\(v) length(v) == (24*3), 
       runner(x=t, k=(24*3), at = seq(0, 24*5, by = 24))) -> OUTPUT

OUTPUT

#> [[1]]
#>  [1]  0  1  2  3  4  5  6  7  8  9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
#> [26] 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49
#> [51] 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71
#> 
#> [[2]]
#>  [1] 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48
#> [26] 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73
#> [51] 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95
#> 
#> [[3]]
#>  [1]  48  49  50  51  52  53  54  55  56  57  58  59  60  61  62  63  64  65  66
#> [20]  67  68  69  70  71  72  73  74  75  76  77  78  79  80  81  82  83  84  85
#> [39]  86  87  88  89  90  91  92  93  94  95  96  97  98  99 100 101 102 103 104
#> [58] 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119

## to get the desired output in seconds 
# OUTPUT * 3600

^{Created on 2025-04-03 with reprex v2.1.1}