The goal of pace is to implement various estimators of Average Causal Effects within principal strata (PACEs) from randomized experiments and observational studies.
devtools::install_github("shuyang1987/pace")Jiang et al. (2020) Multiply robust estimation of causal effects under principal ignorability.
pace(X,Z,S,Y,family.Y,nboot)
| Argument | |
|---|---|
| X | is a matrix of pre-treatment covariates without intercept (n x p). |
| Z | is a vector of treatment (n x 1). |
| S | is a vector of binary intermediate outcome (n x 1). |
| Y | is a vector of outcome ( n x 1). |
| family.Y | specifies the family for the outcome model. |
| * | “gaussian”: a linear regression model for the continuous outcome. |
| * | “binomial”: a logistic regression model for the binary outcome. |
| nboot | is the number of bootstrap samples. |
| Complier | |
| tau10w | a principal score weighting estimator of tau10 |
| tau10sw | a stabilized weighting estimator of tau10 |
| tau10reg | a regression estimator of tau10 using outcome mean regression and inverse probability of treatment weighting |
| tau10reg2 | a regression estimator of tau10 using outcome mean and principal score regression |
| tau10aw | a triply robust estimator of tau10 |
| bootstrap variance estimator | ve.tau10w,ve.tau10sw,ve.tau10reg,ve.tau10reg2,ve.tau10aw |
| Never Taker | |
| tau00w | a principal score weighting estimator of tau00 |
| tau00sw | a stabilized weighting estimator of tau00 |
| tau00reg | a regression estimator of tau00 using outcome mean regression and inverse probability of treatment weighting |
| tau00reg2 | a regression estimator of tau00 using outcome mean and principal score regression |
| tau00aw | a triply robust estimator of tau00 |
| bootstrap variance estimator | ve.tau00w,ve.tau00sw,ve.tau00reg,ve.tau00reg2,ve.tau00aw |
| Always Taker | |
| tau11w | a principal score weighting estimator of tau11 |
| tau11sw | a stabilized weighting estimator of tau11 |
| tau11reg | a regression estimator of tau11 using outcome mean regression and inverse probability of treatment weighting |
| tau11reg2 | a regression estimator of tau11 using outcome mean and principal score regression |
| tau11aw | a triply robust estimator of tau11 |
| bootstrap variance estimator | ve.tau11w,ve.tau11sw,ve.tau11reg,ve.tau11reg2,ve.tau11aw |
Application: An encouragement experiment with non-compliance
fl = read.table("fludata.txt", header=TRUE, quote="\"")
Z = fl$assign
S = fl$receive
Y = fl$outcome
X = as.matrix(fl[, -c(1, 2, 3)])
N = length(Z)
out.flu<-pace::pace(X,Z,S,Y,family.Y="binomial",nboot=100)
cname<-c("Est","SE","95%CI","")
rname<-c("w","sw","reg","reg2")
table10<-rbind(
c(out.flu$tau10w, sqrt(out.flu$ve.tau10w) , out.flu$tau10w-1.96* sqrt(out.flu$ve.tau10w), out.flu$tau10w+1.96*sqrt(out.flu$ve.tau10w) ),
c(out.flu$tau10sw, sqrt(out.flu$ve.tau10sw) , out.flu$tau10sw-1.96* sqrt(out.flu$ve.tau10sw), out.flu$tau10sw+1.96*sqrt(out.flu$ve.tau10sw) ),
c(out.flu$tau10reg, sqrt(out.flu$ve.tau10reg) , out.flu$tau10reg-1.96* sqrt(out.flu$ve.tau10reg), out.flu$tau10reg+1.96*sqrt(out.flu$ve.tau10reg) ),
c(out.flu$tau10reg2, sqrt(out.flu$ve.tau10reg2) , out.flu$tau10reg2-1.96* sqrt(out.flu$ve.tau10reg2), out.flu$tau10reg2+1.96*sqrt(out.flu$ve.tau10reg2) )
)
colnames(table10)<-cname
row.names(table10)<-rname
table00<-rbind(
c(out.flu$tau00w, sqrt(out.flu$ve.tau00w) , out.flu$tau00w-1.96* sqrt(out.flu$ve.tau00w), out.flu$tau00w+1.96*sqrt(out.flu$ve.tau00w) ),
c(out.flu$tau00sw, sqrt(out.flu$ve.tau00sw) , out.flu$tau00sw-1.96* sqrt(out.flu$ve.tau00sw), out.flu$tau00sw+1.96*sqrt(out.flu$ve.tau00sw) ),
c(out.flu$tau00reg, sqrt(out.flu$ve.tau00reg) , out.flu$tau00reg-1.96* sqrt(out.flu$ve.tau00reg), out.flu$tau00reg+1.96*sqrt(out.flu$ve.tau00reg) ),
c(out.flu$tau00reg2, sqrt(out.flu$ve.tau00reg2) , out.flu$tau00reg2-1.96* sqrt(out.flu$ve.tau00reg2), out.flu$tau00reg2+1.96*sqrt(out.flu$ve.tau00reg2) )
)
colnames(table00)<-cname
row.names(table00)<-rname
table11<-rbind(
c(out.flu$tau11w, sqrt(out.flu$ve.tau11w) , out.flu$tau11w-1.96* sqrt(out.flu$ve.tau11w), out.flu$tau11w+1.96*sqrt(out.flu$ve.tau11w) ),
c(out.flu$tau11sw, sqrt(out.flu$ve.tau11sw) , out.flu$tau11sw-1.96* sqrt(out.flu$ve.tau11sw), out.flu$tau11sw+1.96*sqrt(out.flu$ve.tau11sw) ),
c(out.flu$tau11reg, sqrt(out.flu$ve.tau11reg) , out.flu$tau11reg-1.96* sqrt(out.flu$ve.tau11reg), out.flu$tau11reg+1.96*sqrt(out.flu$ve.tau11reg) ),
c(out.flu$tau11reg2, sqrt(out.flu$ve.tau11reg2) , out.flu$tau11reg2-1.96* sqrt(out.flu$ve.tau11reg2), out.flu$tau11reg2+1.96*sqrt(out.flu$ve.tau11reg2) )
)
colnames(table11)<-cname
row.names(table11)<-rname
library(knitr)
print(kable( table10, caption="Results for compliers" ))
#>
#>
#> Table: Results for compliers
#>
#> Est SE 95%CI
#> ----- ----------- ---------- ----------- ----------
#> w -0.0166239 0.0191541 -0.0541660 0.0209183
#> sw -0.0162693 0.0190716 -0.0536496 0.0211111
#> reg -0.0152307 0.0190876 -0.0526424 0.0221809
#> reg2 -0.0149633 0.0191615 -0.0525198 0.0225931
print(kable( table00, caption="Results for never takers" ))
#>
#>
#> Table: Results for never takers
#>
#> Est SE 95%CI
#> ----- ----------- ---------- ----------- ----------
#> w -0.0061187 0.0119362 -0.0295136 0.0172762
#> sw -0.0061003 0.0119541 -0.0295303 0.0173297
#> reg -0.0059885 0.0118392 -0.0291934 0.0172164
#> reg2 -0.0060646 0.0118694 -0.0293285 0.0171994
print(kable( table11, caption="Results for always takers" ))
#>
#>
#> Table: Results for always takers
#>
#> Est SE 95%CI
#> ----- ----------- ---------- ----------- ----------
#> w -0.0466259 0.0244611 -0.0945697 0.0013179
#> sw -0.0465378 0.0245400 -0.0946361 0.0015605
#> reg -0.0472388 0.0244514 -0.0951636 0.0006860
#> reg2 -0.0461420 0.0243074 -0.0937846 0.0015006