causaldrf (version 0.4.2)

wtrg_est: The weighted regression estimator

Description

This method uses weight matrices to estimate parameters for an ADRF with quadratic or linear fits.

Usage

wtrg_est(Y,
         treat,
         covar_formula,
         data,
         e_treat_1,
         e_treat_2,
         e_treat_3,
         e_treat_4,
         degree)

Value

wtrg_est returns an object of class "causaldrf", a list that contains the following components:

param

the estimated parameters.

call

the matched call.

Arguments

Y

is the output

treat

is the treatment variable

covar_formula

is the formula for the covariates model of the form: ~ X.1 + ....

data

will contain all the data: X, treat, and Y

e_treat_1

is estimated treatment

e_treat_2

is estimated treatment squared

e_treat_3

is estimated treatment cubed

e_treat_4

is estimated treatment to the fourth

degree

is 1 for linear fit and 2 for quadratic fit

Details

This function estimates the ADRF by the method described in Schafer and Galagate (2015) which uses weight matrices to adjust for possible bias.

References

Schafer, J.L., Galagate, D.L. (2015). Causal inference with a continuous treatment and outcome: alternative estimators for parametric dose-response models. Manuscript in preparation.

See Also

iptw_est, ismw_est, reg_est, aipwee_est, wtrg_est, etc. for other estimates.

t_mod, overlap_fun to prepare the data for use in the different estimates.

Examples

## Example from Schafer (2015).

example_data <- sim_data


t_mod_list <- t_mod(treat = T,
              treat_formula = T ~ B.1 + B.2 + B.3 + B.4 + B.5 + B.6 + B.7 + B.8,
              data = example_data,
              treat_mod = "Normal")

cond_exp_data <- t_mod_list$T_data
full_data <- cbind(example_data, cond_exp_data)

wtrg_list <- wtrg_est(Y = Y,
                      treat = T,
                      covar_formula = ~ B.1 + B.2 + B.3 + B.4 + B.5 + B.6 + B.7 + B.8,
                      data = example_data,
                      e_treat_1 = full_data$est_treat,
                      e_treat_2 = full_data$est_treat_sq,
                      e_treat_3 = full_data$est_treat_cube,
                      e_treat_4 = full_data$est_treat_quartic,
                      degree = 1)

sample_index <- sample(1:1000, 100)

plot(example_data$T[sample_index],
      example_data$Y[sample_index],
      xlab = "T",
      ylab = "Y",
      main = "weighted regression estimate")

abline(wtrg_list$param[1],
        wtrg_list$param[2],
        lty = 2,
        lwd = 2,
        col = "blue")

legend('bottomright',
        "weighted regression estimate",
        lty = 2,
        lwd = 2,
        col = "blue",
        bty='Y',
        cex=1)

rm(example_data, t_mod_list, cond_exp_data, full_data, wtrg_list, sample_index)