High_Dim_Cov_Adj

lm_lin_example.R

library(estimatr)
library(fabricatr)
library(DeclareDesign)

set.seed(12345)
dat <-
    fabricate(
        N = 50,
        x = sample(1:40, N, replace = TRUE),
        xF = factor(x),
        potential_outcomes(Y ~ 0.1 * Z + x + rnorm(N)),
        Z = sample(0:1, N, replace = TRUE),
        Y = reveal_outcomes(Y ~ Z)
    )


des <- declare_population(dat) +
    declare_assignment(Z=complete_ra(N=N))+
    declare_potential_outcomes(Y ~ 0.1 * Z + x + rnorm(N)) +
    declare_inquiry(ATE=mean(Y_Z_1 - Y_Z_0))+
    declare_reveal(Y,Z) +
    declare_estimator(Y~Z,
        .method = lm_robust,
        .summary = tidy,
        term = "Z",
        inquiry = "ATE",
        label = "lm_unbiased") +
    declare_estimator(Y~Z,covariates=~xF,
        .method = lm_lin,
        .summary = tidy,
        term = "Z",
        inquiry = "ATE",
        label = "lm_lin") +
    declare_estimator(Y~Z+xF,
        .method = lm_robust,
        .summary = tidy,
        term = "Z",
        inquiry = "ATE",
        label = "lm_dummies") +
    declare_estimator(Y~Z*xF,
        .method = lm_robust,
        .summary = tidy,
        term = "Z",
        inquiry = "ATE",
        label = "lm_interact")

dat2 <- draw_data(des)

draw_estimand(des)
draw_estimates(des)
names(des)

## Just vary assignment:
diag <- diagnose_design(des,bootstrap_sims = 0,sims=c(1,1000,1,1,1,1,1,1,1))

diag


lm_lin(Y ~ Z, covariates = ~as.factor(x), data = dat)
lm_robust(Y ~ Z + as.factor(x), data = dat)
summary(lm(Y ~ Z*as.factor(x), data = dat))