Learn R Programming

interferenceCI (version 1.1)

sample.n: Targeted sampling of sharp null hypotheses

Description

Fills in missingness in $\vec{y}(z;\alpha_{s})$ for $z,s=0,1$ based on targeted sampling algorithm described in Section 4.2 of Rigdon and Hudgens (2014)

Usage

sample.n(eff, y0.a0, y1.a0, y0.a1, y1.a1, p00, p10, p01, p11, n, m.a0, m.a1)

Arguments

eff
treatment effect of interest; either ``DEa0'', ``DEa1'', ``IE'', ``TE'', or ``OE''
y0.a0
Observed $\vec{y}(0;\alpha_{0})$; includes NAs where missing
y1.a0
Observed $\vec{y}(1;\alpha_{0})$; includes NAs where missing
y0.a1
Observed $\vec{y}(0;\alpha_{1})$; includes NAs where missing
y1.a1
Observed $\vec{y}(1;\alpha_{1})$; includes NAs where missing
p00
Missingness in $\vec{y}(0;\alpha_{0})$ is filled in by sampling from a Bernoulli distribution with mean $p_{00}$
p10
Missingness in $\vec{y}(1;\alpha_{0})$ is filled in by sampling from a Bernoulli distribution with mean $p_{10}$
p01
Missingness in $\vec{y}(0;\alpha_{1})$ is filled in by sampling from a Bernoulli distribution with mean $p_{01}$
p11
Missingness in $\vec{y}(1;\alpha_{0})$ is filled in by sampling from a Bernoulli distribution with mean $p_{11}$
n
group size vector where element $i=1,\ldots,k$ is equal to the number of subjects in group $i$
m.a0
$\alpha_{0}$ randomization vector where element $i=1,\ldots,k$ is equal to the number of subjects in group $i$ who would receive treatment if group $i$ was randomized to strategy $\alpha_{0}$
m.a1
$\alpha_{1}$ randomization vector where element $i=1,\ldots,k$ is equal to the number of subjects in group $i$ who would receive treatment if group $i$ was randomized to strategy $\alpha_{1}$

Value

y0.a0
value of $\vec{y}(0;\alpha_{0})$ after missingness has been filled in using targeted sampling
y1.a0
value of $\vec{y}(1;\alpha_{0})$ after missingness has been filled in using targeted sampling
y0.a1
value of $\vec{y}(0;\alpha_{1})$ after missingness has been filled in using targeted sampling
y1.a1
value of $\vec{y}(1;\alpha_{1})$ after missingness has been filled in using targeted sampling
effect
value of treatment effect of interested under sharp null after missingness filled in using targeted sampling

References

Rigdon, J. and Hudgens, M.G. ``Exact confidence intervals in the presence of interference.'' Submitted to Statistics and Probability Letters 2014.

See Also

exactCI