tests/testthat/test_sunab.R

In fwildclusterboot: Fast Wild Cluster Bootstrap Inference for Linear Models

test_that("test sunab", {

  skip_on_cran()
  skip_if_not(
    find_proglang("julia"), 
    message = "skip test as julia installation not found."
  )
  
  library(fixest)
  # Simple DiD example
  base_stagg <- fixest::base_stagg

  # The DiD estimation
  res_sunab <- fixest::feols(
    y ~ x1 + fixest:::sunab(year_treated, year) | id + year, base_stagg)
  
  res_sunab_3ref <- fixest::feols(
    y ~ x1 + fixest:::sunab(year_treated, year, ref.p = c(.F + 0:2, -1)) |
                           id + year, 
                         cluster = "id", base_stagg, 
                         ssc = fixest::ssc(adj = FALSE, cluster.adj = FALSE))
  
  fixest_att <- aggregate(res_sunab, agg = "ATT")
  fixest_ci <- as.matrix(confint(summary(res_sunab, agg = "ATT")))
  fixest_pval <- as.vector(pvalue(summary(res_sunab, agg = "ATT")))
  
  
  # test ATT equivalence
  boot_att <- 
  boot_aggregate(
    res_sunab, 
    B = 99999, 
    agg = "ATT", 
    clustid = "id", 
    ssc = boot_ssc(adj = FALSE, cluster.adj = FALSE)
  )
  
  expect_equal(
    fixest_ci["ATT",],
    boot_att[,3:4], 
    ignore_attr = TRUE,
    tolerance = 0.03
    )
  
  expect_equal(
    fixest_pval[2],
    boot_att[,2], 
    ignore_attr = TRUE, 
    tolerance = 0.01
  )
  
  
  # effects aggregated by year
  fixest_ci <- as.matrix(confint(summary(res_sunab, agg = TRUE)))
  fixest_pval <- as.vector(pvalue(summary(res_sunab, agg = TRUE)))
  
  boot_year <- 
    boot_aggregate(
      res_sunab, 
      B = 99999, 
      agg = TRUE, 
      clustid = "id", 
      ssc = boot_ssc(adj = FALSE, cluster.adj = FALSE)
    )
  
  expect_equal(
    fixest_ci[-1,],
    boot_year[,3:4], 
    ignore_attr = TRUE, 
    tolerance = 0.05
  )
  
  expect_equal(
    fixest_pval[-1],
    boot_year[,2], 
    ignore_attr = TRUE, 
    tolerance = 0.05
  )
  
  

})