Store regression coefs in R data.table

Question

I would like to run a regression and store the coefficients in a data.table. Here is a minimal example:

library(data.table)
library(MASS)

dt <- as.data.table(iris)
dt[, c("coef1", "coef2") := rlm(Sepal.Length ~ Petal.Length)$coef]
dt

However it recycles the output so the coef1 and coef2 are the same on each line, but the two coefficients show up on every other line.

This example shows how I would like it to look, but it is not optimal because it requires running the regression twice:

dt <- as.data.table(iris)
dt[, `:=`("coef1"=rlm(Sepal.Length ~ Petal.Length)$coef[1], "coef2"=rlm(Sepal.Length ~ Petal.Length)$coef[2])]
dt

nghauran nghauran · Accepted Answer · 2018-09-05T17:35:42

It looks like you want to store by row. Try this:

library(data.table)
library(MASS)
dt <- as.data.table(iris)
dt[, c("coef1", "coef2") := as.list(rlm(Sepal.Length ~ Petal.Length)$coef)][]
dt
# output
     Sepal.Length Sepal.Width Petal.Length Petal.Width   Species    coef1     coef2
  1:          5.1         3.5          1.4         0.2    setosa 4.300878 0.4097214
  2:          4.9         3.0          1.4         0.2    setosa 4.300878 0.4097214
  3:          4.7         3.2          1.3         0.2    setosa 4.300878 0.4097214
  4:          4.6         3.1          1.5         0.2    setosa 4.300878 0.4097214
  5:          5.0         3.6          1.4         0.2    setosa 4.300878 0.4097214
 ---                                                                               
146:          6.7         3.0          5.2         2.3 virginica 4.300878 0.4097214
147:          6.3         2.5          5.0         1.9 virginica 4.300878 0.4097214
148:          6.5         3.0          5.2         2.0 virginica 4.300878 0.4097214
149:          6.2         3.4          5.4         2.3 virginica 4.300878 0.4097214
150:          5.9         3.0          5.1         1.8 virginica 4.300878 0.4097214

Store regression coefs in R data.table

1 Answers