cfa: cfa

In ccfa: Continuous Counterfactual Analysis

Description

compute counterfactuals using distribution regression with a continuous treatment

Usage

cfa(formla, xformla = NULL, tvals, yvals, data, method = "dr",
  link = "logit", tau = seq(0.01, 0.99, 0.01), condDistobj = NULL,
  se = TRUE, iters = 100, cl = 1)

Arguments

formla	a formula y ~ treatment
xformla	one sided formula for x variables to include, e.g. ~x1 + x2
tvals	the values of the "treatment" to compute parameters of interest for
yvals	the values to compute the counterfactual distribution for
data	the data.frame where y, t, and x are
method	either "dr" or "qr" for distribution regression or quantile regression
link	if using distribution regression, any link function that works with the binomial family (e.g. logit (the default), probit, cloglog)
tau	if using quantile regression, which values of tau to estimate the conditional quantiles
condDistobj	optional conditional distribution object that has been previously computed
se	whether or not to compute standard errors using the bootstrap
iters	how many bootstrap iterations to use
cl	how many clusters to use for parallel computation of standard errors

Value

CFA object

Examples

data(igm)
tvals <- seq(10,12,length.out=8)
yvals <- seq(quantile(igm$lcfincome, .05), quantile(igm$lcfincome, .95), length.out=50)
## This line doesn't adjust for any covariates
cfa(lcfincome ~ lfincome, tvals=tvals, yvals=yvals, data=igm,
 se=FALSE)

## This line adjusts for differences in education
cfa(lcfincome ~ lfincome, ~HEDUC, tvals=tvals, yvals=yvals, data=igm,
 se=FALSE)

ccfa documentation built on May 2, 2019, 7:28 a.m.