Skip to contents

Formula: nu_g = (1/|g|) sum_{l in g} sum_e CATT(e,l) Pr{E=e | E in [-l, T-l]}

Usage

Iwdid(y, unit, time, cohort, never = 0)

Arguments

y

See Usage.

unit

See Usage.

time

See Usage.

cohort

See Usage.

never

See Usage.

Value

List with event_time, att, overall, cohorts, n.

References

Sun and Abraham (2021), Estimating dynamic treatment effects in event studies with heterogeneous treatment effects, Journal of Econometrics 225(3):175-199. Equations (26) and (28) for the estimand and the IW estimator, Section 4.2 for the DID choice of pre-period and control cohort. Verified against the paper.

Examples

Iwdid(y = c(1, 2, 3, 4, 5, 6, 7, 8), unit = c(1, 2, 3, 4, 5, 6, 7, 8),
  time = c(1, 2, 3, 4, 5, 6, 7, 8), cohort = c(1, 2, 3, 4, 5, 6, 7, 8))
#> $event_time
#> numeric(0)
#> 
#> $att
#> numeric(0)
#> 
#> $overall
#> [1] NA
#> 
#> $cohorts
#> [1] 1 2 3 4 5 6 7 8
#> 
#> $n
#> [1] 8
#> 
#> $method
#> [1] "Sun-Abraham interaction-weighted DID"
#> 
#> attr(,"class")
#> [1] "morie_rich_result" "list"