Skip to contents

Borusyak, Jaravel & Spiess (2024) imputation estimator

Source: R/did_modern_native.R
morie_did_borusyak.Rd

Fits unit and time fixed effects on the UNTREATED observations only (via the native alternating demeaner), imputes the untreated potential outcome for every treated cell, and averages the treatment-minus-imputation differences. Standard error by cluster (unit) bootstrap of the whole impute-and-average pipeline – conservative and assumption-light.

Usage

morie_did_borusyak(
  data,
  outcome,
  unit,
  time,
  treatment_time,
  n_bootstrap = 199L,
  seed = 42L,
  alpha = 0.05
)

Arguments

data

Long panel data frame.

outcome, unit, time

Column names.

treatment_time

First treated period per unit (NA = never).

n_bootstrap

Cluster-bootstrap replications. Default 199.

seed

RNG seed.

alpha

CI tail.

Value

List of class "morie_did": estimate, std.error, conf.int, p.value, method, n_units, n_periods, call.

References

Borusyak, Jaravel & Spiess (2024) REStud 91(6).

Examples

df <- expand.grid(id = 1:40, t = 1:8)
df$g <- ifelse(df$id <= 20, 5L, NA)
df$y <- rnorm(nrow(df)) + ifelse(!is.na(df$g) & df$t >= df$g, 2, 0)
morie_did_borusyak(df, "y", "id", "t", "g", n_bootstrap = 29L)
#> Difference-in-differences -- Borusyak-Jaravel-Spiess imputation (cluster bootstrap SE) 
#>   ATT: 2.3454  (SE 0.1700)  95% CI [2.0123, 2.6785]  p = 2.55e-43
#>   Units: 40  Periods: 8