sen.ci: Point estimate and confidence interval for sensitivity...

In qingyuanzhao/CrossScreening: Cross-Screening in Observational Studies that Test Many Hypotheses

Description

Point estimate and confidence interval for sensitivity analysis

Usage

sen.ci(d, mm = c(2, 2, 2), gamma = 1, alpha = 0.05, alpha.up = alpha/2,
  alpha.low = alpha/2, score.method = c("approximate", "exact"))

Arguments

d	a vector of treatment-minus-control differences
mm	a vector (m, munder, mover) that indicates the U-statistic. Does not support matrix mm in this function.
gamma	a vector of sensitivity parameters (must be >= 1).
alpha	significance level for the outer confidence interval
alpha.up	upper-tail probability of the confidence interval
alpha.low	lower-tail probability of the confidence interval
score.method	either approximate score or exact score

Details

See the senmwCI function in the sensitivitymw package.

Value

a list

point.estimate

An interval of point estimates allowing for a bias of gamma in treatment assignment.

ci

An confidence interval allowing for a bias of gamma in treatment assignment.

Author(s)

Qingyuan Zhao

Examples

data(lead)
d.lead <- lead$exposed - lead$control
sen.ci(d.lead, gamma = c(1, 2), alpha.up = 0, alpha.low = 0.05)

qingyuanzhao/CrossScreening documentation built on May 26, 2019, 11:35 a.m.