Estimates the canonical two-group / two-period DiD treatment effect $$\hat\tau = (\bar Y_{1,\text{post}} - \bar Y_{1,\text{pre}}) - (\bar Y_{0,\text{post}} - \bar Y_{0,\text{pre}}).$$ With covariates, fits the regression \(Y = \alpha + \beta D + \gamma P + \tau (D \times P) + X\delta + \varepsilon\) and reports \(\hat\tau\).
Usage
morie_did_2x2(
data,
outcome,
treatment,
post,
covariates = NULL,
cluster = NULL,
alpha = 0.05
)Arguments
- data
A data frame containing the outcome, treatment, post and any covariate columns.
- outcome
Name of the outcome column.
- treatment
Name of the binary (0/1) treatment-group column.
- post
Name of the binary (0/1) post-period column.
- covariates
Optional character vector of covariate column names.
- cluster
Optional cluster ID column for CR1 standard errors.
- alpha
Significance level for confidence intervals (default 0.05).
Value
A list with elements estimate, std_error,
t_stat, p_value, ci_lower, ci_upper,
n_treated, n_control, method, details.
Details
For multi-period staggered designs prefer
morie_did_group_time_att (Callaway-Sant'Anna via
native). morie_did_doubly_robust is the recommended
option when pre-treatment covariates are available.
References
Angrist, J. D., & Pischke, J.-S. (2009). Mostly Harmless Econometrics. Princeton University Press.
Examples
# \donttest{
df <- data.frame(
y = rnorm(200),
d = rep(c(0, 1), each = 100),
post = rep(c(0, 1), times = 100)
)
morie_did_2x2(df, "y", "d", "post")
#> $estimate
#> [1] -0.2594501
#>
#> $std_error
#> interaction
#> 0.2744695
#>
#> $t_stat
#> interaction
#> -0.9452787
#>
#> $p_value
#> interaction
#> 0.3445166
#>
#> $ci_lower
#> interaction
#> -0.7974004
#>
#> $ci_upper
#> interaction
#> 0.2785002
#>
#> $n_treated
#> [1] 100
#>
#> $n_control
#> [1] 100
#>
#> $method
#> [1] "did_2x2"
#>
#> $details
#> $details$all_coefficients
#> [1] 0.03040399 0.05675479 0.16807438 -0.25945014
#>
#> $details$all_se
#> (Intercept) d p interaction
#> 0.1288547 0.1749138 0.1937529 0.2744695
#>
#> $details$n_obs
#> [1] 200
#>
#>
# }