Partial out controls — idid_partial

Compute the partialled out version of a variable based on the regression model provided.

Usage

idid_partial_out(reg, var_to_partial, var_interest = var_to_partial, ...)

Arguments

reg: A regression object.
var_to_partial: A string. The name of the variable for which to compute a partialled out version.
var_interest: A string. The name of the main variable of interest.
...: Additional elements to pass to the regression function when partialling out controls.

Value

A vector of the partialled out version of var_to_partial.

Details

The partialling out procedure is implemented using the same estimation method as the one used in reg.

If var_interest is specified (and different from var_to_partial), var_interest is not partialled out. This is espetially interesting if one wants to partial out controls but not the variable of interest.

Examples

# example with a lm regression
reg_ex_lm <- ggplot2::txhousing |>
  lm(formula = log(sales) ~ median + listings + city + as.factor(date))

idid_partial_out(reg_ex_lm, "median") |>
 head()
#>         1         2         3         4         5         6 
#> 10160.214 -3303.202 -4227.209  3889.547  2092.269 -3574.464 

idid_partial_out(reg_ex_lm, "log(sales)", "median") |>
 head()
#>            1            2            3            4            5            6 
#> -0.050652563 -0.045605122  0.005417663 -0.160510848 -0.033033531  0.002041210 

# example with a fixest regression
reg_ex_fixest <- ggplot2::txhousing  |>
  fixest::feols(fml = log(sales) ~ median + listings |  as.factor(date) + city)
#> NOTE: 1,434 observations removed because of NA values (LHS: 568, RHS: 1,434).

idid_partial_out(reg_ex_fixest, "median") |>
  head()
#> NOTE: 1,434 observations removed because of NA values (LHS: 616, RHS: 1,424).
#> [1] 10160.214 -3303.202 -4227.209  3889.547  2092.269 -3574.464

idid_partial_out(reg_ex_fixest, "log(sales)", "median") |>
  head()
#> NOTE: 1,426 observations removed because of NA values (LHS: 568, RHS: 1,424).
#> [1] -0.050652563 -0.045605122  0.005417663 -0.160510848 -0.033033530
#> [6]  0.002041210