RDTEfficiencyBound: Finite-sample efficiency bounds for minimax CIs

In kolesarm/RDHonest: Honest Inference in Regression Discontinuity Designs

Finite-sample efficiency bounds for minimax CIs

Description

Compute efficiency of minimax one-sided CIs at constant functions, or efficiency of two-sided fixed-length CIs at constant functions under second-order Taylor smoothness class.

Usage

RDTEfficiencyBound(object, opt.criterion = "FLCI", beta = 0.5)

Arguments

object	An object of class "RDResults", typically a result of a call to RDHonest.
opt.criterion	Either "FLCI" for computing efficiency of two-sided CIs, or else "OCI" for minimax one-sided CIs.
beta	Determines quantile of excess length for evaluating minimax efficiency of one-sided CIs. Ignored if opt.criterion=="FLCI".

Value

Efficiency bound, a numeric vector of length one.

References

Timothy B. Armstrong and Michal Kolesár. Optimal inference in a class of regression models. Econometrica, 86(2):655–683, March 2018. \Sexpr[results=rd]{tools:::Rd_expr_doi("10.3982/ECTA14434")}

Examples

r <- RDHonest(voteshare ~ margin, data=lee08,
              subset=abs(margin)<10, M=0.1, h=2)
RDTEfficiencyBound(r, opt.criterion="OCI")

kolesarm/RDHonest documentation built on April 14, 2024, 3:27 a.m.