estlate: Estimation of local average treatment effects under...
In ri: ri: R package for performing randomization-based inference for experiments

Description Usage Arguments Value Note Author(s) References See Also Examples

Function for estimating the local average treatment effect (LATE) via variants of Wald/2SLS estimation (taking the ratio of two average treatment effect estimators). Permits regression adjustment for covariates, difference estimation (with a pretreatment measure of the outcome variable), inverse probability weighting and Horvitz-Thompson estimation.

1	estlate(Y, D, Z, X = NULL, Ypre = NULL, Dpre = NULL, prob = NULL, HT = FALSE)

`Y`	numeric vector of length N, outcome variable
`D`	binary vector (0 or 1) of length N, treatment receipt indicator
`Z`	binary vector (0 or 1) of length N, treatment assignment indicator
`X`	N-by-k numeric matrix of covariates for regression adjustment
`Ypre`	numeric vector of length N, pretreatment measure of the outcome variable for difference estimation
`Dpre`	numeric vector of length N, pretreatment measure of the treatment receipt variable for difference estimation
`prob`	numeric vector within the (0,1) interval of N-length, probability of treatment assignment, as output by `genprob()` or `genprobexact()`. When `prob=NULL` (by default), assumes 0.5 probability of assignment to treatment
`HT`	when `HT=TRUE`, invokes the Horvitz-Thompson (difference-in-totals) estimator. When `HT=FALSE`, invokes the inverse-probability-weighted regression estimator

a numeric scalar, the estimated LATE

Takes the ratio of two estate values, the numerator with Y as the outcome variable and Z as the treatment indicator, the denominator with D as the outcome variable and Z as the treatment indicator

Peter M. Aronow <peter.aronow@yale.edu>; Cyrus Samii <cds2083@nyu.edu>

Angrist, Joshua D, Guido W. Imbens and Donald B. Rubin. 1996. Identification of Causal Effects Using Instrumental Variables. J. Amer. Statist. Assoc. 91: 444-55.

Gerber, Alan S. and Donald P. Green. 2012. Field Experiments: Design, Analysis, and Interpretation. New York: W.W. Norton.

Horvitz, D.G. and D.J. Thompson. 1952. A generalization of sampling without replacement from a finite universe. J. Amer. Statist. Assoc. 47: 663-684.

estate

y <- c(8,6,2,0,3,1,1,1,2,2,0,1,0,2,2,4,1,1) 
Z <- c(1,1,0,0,1,1,0,0,1,1,1,1,0,0,1,1,0,0)
D <- c(1,0,0,0,0,0,1,0,1,1,0,1,0,0,1,0,0,1)

cluster <- c(1,1,2,2,3,3,4,4,5,5,6,6,7,7,8,8,9,9)
block <- c(rep(1,4),rep(2,6),rep(3,8))
probs <- genprobexact(Z,block,cluster) # generate probability of assignment
late <- estlate(y,D,Z,prob=probs) # estimate the LATE; estimated LATE = 9