gam_est: The GAM estimator

In causaldrf: Estimating Causal Dose Response Functions

The GAM estimator

Description

This estimates the ADRF using a method similar to that described in Hirano and Imbens (2004), but with spline basis terms in the outcome model.

Usage

gam_est(Y,
        treat,
        treat_formula,
        data,
        grid_val,
        treat_mod,
        link_function,
        ...)

Arguments

Y	is the the name of the outcome variable contained in data.
treat	is the name of the treatment variable contained in data.
treat_formula	an object of class "formula" (or one that can be coerced to that class) that regresses treat on a linear combination of X: a symbolic description of the model to be fitted.
data	is a dataframe containing Y and treat and X.
grid_val	contains the treatment values to be evaluated.
treat_mod	a description of the error distribution to be used in the model for treatment. Options include: "Normal" for normal model, "LogNormal" for lognormal model, "Sqrt" for square-root transformation to a normal treatment, "Poisson" for Poisson model, "NegBinom" for negative binomial model, "Gamma" for gamma model.
link_function	is either "log", "inverse", or "identity" for the "Gamma" treat_mod.
...	additional arguments to be passed to the gam() outcome function.

Details

This function estimates the ADRF similarly to the method described by Hirano and Imbens (2004), but with a generalized additive model in the outcome model.

Value

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

param	parameter estimates for a gam fit.
t_mod	the result of the treatment model fit.
out_mod	the result of the outcome model fit.
call	the matched call.

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.

Hirano, Keisuke, Imbens, Guido W (2004). The propensity score with continuous treatments. Applied Bayesian modeling and causal inference from incomplete-data perspectives.

Flores, Carlos A and Flores-Lagunes, Alfonso and Gonzalez, Arturo and Neumann, Todd C (2012). Estimating the effects of length of exposure to instruction in a training program: the case of job corps. Review of Economics and Statistics. 94.1, 153-171

See Also

nw_est, iw_est, hi_est, gam_est, add_spl_est, bart_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

gam_list <- gam_est(Y = Y,
                treat = T,
                treat_formula = T ~ B.1 + B.2 + B.3 + B.4 + B.5 + B.6 + B.7 + B.8,
                data = example_data,
                grid_val = seq(8, 16, by = 1),
                treat_mod = "Normal")

sample_index <- sample(1:1000, 100)

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

lines(seq(8, 16, by = 1),
      gam_list$param,
      lty = 2,
      lwd = 2,
      col = "blue")

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

rm(example_data, gam_list, sample_index)

causaldrf documentation built on Sept. 30, 2022, 1:07 a.m.