Equation 3.4 of FEDAI. The standard error depends on the variance of both
schedules of potential outcomes and on their covariance, which is why it
cannot be computed from observed data: no unit reveals both of its potential
outcomes.
Usage
equation_3.4(var_Y0, var_Y1, cov_Y0_Y1, N, m)
Arguments
- var_Y0
the variance of the untreated potential outcomes
- var_Y1
the variance of the treated potential outcomes
- cov_Y0_Y1
the covariance of the treated and untreated potential outcomes
- N
the number of units in the experiment
- m
the number of units assigned to treatment
Value
A numeric scalar, the standard error of the difference-in-means
estimator.
Examples
var_Y0 <- pop_var(table_2_1$Y_Z_0)
var_Y1 <- pop_var(table_2_1$Y_Z_1)
cov_Y0_Y1 <- pop_cov(table_2_1$Y_Z_0, table_2_1$Y_Z_1)
equation_3.4(var_Y0 = var_Y0,
var_Y1 = var_Y1,
cov_Y0_Y1 = cov_Y0_Y1,
N = 7,
m = 2)
#> [1] 4.603311