CODEWITHSUNDEEP
rdplyrsubset
Nick Knauer
Nick Knauer

Reputation: 4243

Select column by value and following 3 columns

I have a data.table like this: 

   Model                Variable   Coefficient standardized coefficient Model                Variable   Coefficient standardized coefficient Model
1:     0               Intercept  7.136994e+12              0.919481694   0.1               Intercept  2.201799e+12              0.918641093   0.2
2:     0 Comp_BK_Tot_Tot050500_i -1.177355e+13             -0.005086289   0.1 Comp_BK_Tot_Tot050500_i -3.632202e+12             -0.001569146   0.2
                  Variable   Coefficient standardized coefficient Model                Variable   Coefficient standardized coefficient
1:               Intercept  2.244410e+12              0.918648351   0.3               Intercept  2.258975e+12              0.918650832
2: Comp_BK_Tot_Tot050500_i -3.702495e+12             -0.001599514   0.3 Comp_BK_Tot_Tot050500_i -3.726523e+12             -0.001609894

Here is my issue, I have a variable saved as "0.3" as new_num. How do I write and expression where if the column name "Model" contains = new_num, then select that column and the next 3 columns to the right of it?

So for example, I want my output in this case to be: 

Model                   Variable     Coefficient     standardized coefficient
  0.3                  Intercept    2.258975e+12                  0.918650832 
  0.3    Comp_BK_Tot_Tot050500_i   -3.726523e+12                 -0.001609894

Upvotes: 2

Views: 40

Answers (1)

akrun
akrun

Reputation: 887118

We can subset the columns first with an index ('i1') and then assign 'Model' with the values of 'new_num'

i1 <- match("Model", names(dt))
dt1 <- dt[, i1:(i1+3), with = FALSE][, Model := new_num][]

Upvotes: 1

Related Questions