unfortunately I have a problem and hope anybody has an idea.
I want to use the first difference for some time series.
I have a Datamatrix with 100 columns. Every is a timeseries.
For example I can create with: v <-diff(DM$column1) a new vector with my wanted result.
But is it possible to take more as one column? For example from $column1 until $column30.
And can I overwrite the columns with the new values directly in the original DM matrix?
Thank you very much for any ideas.
Best regards
You can use apply to apply the function diff to each column.
Let's take an example matrix representing 5 time series
set.seed(1)
m <- `colnames<-`(replicate(5, rnorm(10)), LETTERS[1:5])
m
#> A B C D E
#> [1,] -0.6264538 1.51178117 0.91897737 1.35867955 -0.1645236
#> [2,] 0.1836433 0.38984324 0.78213630 -0.10278773 -0.2533617
#> [3,] -0.8356286 -0.62124058 0.07456498 0.38767161 0.6969634
#> [4,] 1.5952808 -2.21469989 -1.98935170 -0.05380504 0.5566632
#> [5,] 0.3295078 1.12493092 0.61982575 -1.37705956 -0.6887557
#> [6,] -0.8204684 -0.04493361 -0.05612874 -0.41499456 -0.7074952
#> [7,] 0.4874291 -0.01619026 -0.15579551 -0.39428995 0.3645820
#> [8,] 0.7383247 0.94383621 -1.47075238 -0.05931340 0.7685329
#> [9,] 0.5757814 0.82122120 -0.47815006 1.10002537 -0.1123462
#> [10,] -0.3053884 0.59390132 0.41794156 0.76317575 0.8811077
Then to diff each column all at once, we can do:
apply(m, 2, diff)
#> A B C D E
#> [1,] 0.8100971 -1.12193793 -0.13684107 -1.46146728 -0.08883808
#> [2,] -1.0192719 -1.01108382 -0.70757132 0.49045934 0.95032506
#> [3,] 2.4309094 -1.59345931 -2.06391668 -0.44147665 -0.14030018
#> [4,] -1.2657730 3.33963081 2.60917744 -1.32325452 -1.24541889
#> [5,] -1.1499762 -1.16986453 -0.67595449 0.96206499 -0.01873946
#> [6,] 1.3078974 0.02874335 -0.09966677 0.02070461 1.07207712
#> [7,] 0.2508957 0.96002647 -1.31495688 0.33497656 0.40395096
#> [8,] -0.1625434 -0.12261502 0.99260233 1.15933877 -0.88087914
#> [9,] -0.8811697 -0.22731987 0.89609162 -0.33684962 0.99345394
Note that when you diff a vector, the result is one element shorter, so this matrix is one row shorter than the original.
