ate: Functions to estimate the average treatment effect

In denisagniel/synthate: Synthetic Causal Inference

Description

Functions to estimate the average treatment effect

Usage

ate(this_data, specific_ate, ...)

naive_ate(this_data, ...)

ipw1_ate(this_data, ...)

ipw2_ate(this_data, ...)

ipw3_ate(this_data, ...)

strat_ate(this_data, ...)

rf_ate(this_data, outcome_fm, ...)

cart_ate(this_data, outcome_fm, ...)

pscart_ate(this_data, ps_fm, ...)

strat_regr_ate(this_data, outcome_fm, outcome_fam = gaussian, ...)

match_ps_ate(this_data, ...)

match_prog_ate(this_data, ...)

match_both_ate(this_data, ...)

caliper_ps_ate(this_data, ...)

caliper_prog_ate(this_data, ...)

caliper_both_ate(this_data, ...)

caliper_nn_ate(this_data, ...)

caliper_logit_ate(this_data, ...)

regr_ate(this_data, outcome_fm, outcome_fam = gaussian, ...)

dr_ate(this_data, outcome_fm, outcome_fam = gaussian, ...)

bal_ate(this_data, cov_ids, ...)

highdim_bal_ate(this_data, cov_ids, ...)

cem_ate(this_data, cov_ids, ...)

grf_ate(this_data, cov_ids, ...)

Arguments

this_data	a data frame with value y for the outcome, d for the treatment, and w for a case weight.
...	additional arguments to be passed indicating, for example, the outcome formula, propensity score formula, or other things on a case by case basis.

Value

a 1 x 1 data frame containing the named ATE

Examples

ate_list <- list(
ipw1 = ipw1_ate,
ipw2 = ipw2_ate,
ipw3 = ipw3_ate,
strat = strat_ate,
strat_regr = strat_regr_ate,
match_ps = match_ps_ate,
match_prog = match_prog_ate,
match_both = match_both_ate,
cal_ps = caliper_ps_ate,
cal_both = caliper_both_ate,
regr = regr_ate,
dr = dr_ate,
bal = bal_ate,
hdbal = highdim_bal_ate
)
gen_mod <- generate_data(n = 100, 
                         dgp = 'ks', 
                         correct_outcome = FALSE,
                         correct_ps = TRUE)
this_data <- gen_mod$data
this_data <- estimate_scores(this_data, outcome_fm = outcome_fm,
                             ps_fm = ps_fm,
                             ps_fam = ps_fam,
                             outcome_fam = outcome_fam)
thetahat <- with(gen_mod,
                 estimate_ates(this_data,
                          ate_list,
                          cov_ids = cov_ids,
                          outcome_fm = stringr::str_c('d + ', outcome_fm),
                          outcome_fam = outcome_fam))
thetahat

denisagniel/synthate documentation built on April 16, 2020, 12:45 a.m.