Parallel to morie_did_synthetic: the same native
Arkhangelsky et al. (2021) SDID engine, surfaced with the
placebo / bootstrap / jackknife variance options and the raw
unit / time weights. Use morie_did_synthetic when you want
the rmorie result-list shape consumed by morie_did_*
downstream code.
Usage
morie_did_synthdid_estimate(
panel,
unit,
time,
treatment,
outcome,
vcov_method = "placebo",
...
)Arguments
- panel
Long-format balanced panel.
- unit
Unit identifier column.
- time
Time period column.
- treatment
Binary (0/1) treatment indicator that turns on at onset for treated units and is zero everywhere for controls.
- outcome
Outcome column.
- vcov_method
Inference method: one of
"placebo"(default),"bootstrap","jackknife".- ...
Ignored; retained for back-compat.
Value
An S3 list of class morie_did_synthdid_result with
elements att, std_error, vcov_method,
n_treated, n_control, n_pre,
n_post, method, and raw (the full
native SDID fit list).
References
Arkhangelsky, D., Athey, S., Hirshberg, D. A., Imbens, G. W., & Wager, S. (2021). Synthetic difference-in-differences. American Economic Review, 111(12), 4088–4118.
Examples
if (requireNamespace("coresynth", quietly = TRUE)) {
set.seed(13)
df <- expand.grid(unit = 1:30, time = 1:8)
df$treat_time <- ifelse(df$unit <= 5, 6, Inf)
df$d <- as.integer(df$time >= df$treat_time)
df$y <- 0.1 * df$time + 0.5 * df$d + rnorm(nrow(df), sd = 0.3)
out <- morie_did_synthdid_estimate(df, unit = "unit", time = "time",
treatment = "d", outcome = "y")
out$att
}
#> [1] 0.2555796