Simple function that replicates fixest objects while (optionally) computing different standard-errors. Useful mostly in combination with etable or coefplot.
# S3 method for fixest
rep(x, times = 1, each = 1, cluster, ...)# S3 method for fixest_list
rep(x, times = 1, each = 1, cluster, ...)
.l(...)
Either a fixest object, either a list of fixest objects created with .l().
Integer vector giving the number of repetitions of the vector of elements. By default times = 1. It must be either of length 1, either of the same length as the argument x.
Integer scalar indicating the repetition of each element. Default is 1.
A list containing the types of standard-error to be computed, default is missing. If not missing, it must be of the same length as times, each, or the final vector. Note that if the arguments times and each are missing, then times becomes equal to the length of cluster. (Note that cluster accepts the character values "standard" or "hetero" to compute non-clustered SEs.)
In .l(): fixest objects. In rep(): not currently used.
Returns a list of the appropriate length. Each element of the list is a fixest object.
To apply rep.fixest on a list of fixest objects, it is absolutely necessary to use .l() and not list().
# NOT RUN {
# Let's show results with different standard-errors
est = feols(Ozone ~ Solar.R + Wind + Temp, data = airquality)
my_cluster = list("Month", "Day", ~ Day + Month)
etable(rep(est, cluster = my_cluster))
coefplot(rep(est, cluster = my_cluster), drop = "Int")
#
# To rep multiple objects, you need to use .l()
#
est_bis = feols(Ozone ~ Solar.R + Wind + Temp | Month, airquality)
etable(rep(.l(est, est_bis), cluster = my_cluster))
# using each
etable(rep(.l(est, est_bis), each = 3, cluster = my_cluster))
# }
Run the code above in your browser using DataLab