RI2by2 (version 1.4)

Robins.CI: Asymptotic confidence interval for a treatment effect on a binary outcome

Description

Computes the Robins (1988) confidence interval for the average treatment effect on a binary outcome in an experiment where \(m\) of \(n\) individuals are randomized to treatment by design.

Usage

Robins.CI(data, level)

Value

tau.hat

estimated average treatment effect

lower

lower bound of confidence interval

upper

upper bound of confidence interval

Arguments

data

observed 2 by 2 table in matrix form where row 1 is the treatment assignment Z=1 and column 1 is the binary outcome Y=1

level

significance level of hypothesis tests, i.e., method yields a 100(1-level)% confidence interval

Author

Joseph Rigdon

Details

The Robins (1988) confidence interval is similar in form to the well known Wald confidence interval for a difference in proportions, but is guaranteed to have smaller width.

References

Robins, J.M. (1988). Confidence intervals for causal parameters. Statistics in Medicine, 7(7), 773-785.

Examples

#Example 1 from Robins (1988)
ex = matrix(c(40,60,15,85),2,2,byrow=TRUE)
Robins.CI(ex,0.05)