imputedata

Imputation results based on Missing At Random(MAR) and Missing Completely At Random(MCAR) assumption for treatment and control group.

datasets

Using the idea of "tipping point" (proposed in
Gregory Campbell, Gene Pennello and Lilly Yue(2011)
<DOI:10.1080/10543406.2011.550094>) to visualize the results
of sensitivity analysis for missing data, the package provides
a set of functions to list out all the possible combinations
of missing values in two treatment arms, calculate
corresponding estimated treatment effects and p values, and draw
a colored heat-map. It could deal with randomized
experiments with a binary outcome or a continuous outcome. In addition,
the package provides a visualized method to compare various imputation
methods by adding the rectangles or convex hulls on the basic plot.

Xikun Han

TippingPoint

Enhanced Tipping Point Displays the Results of Sensitivity
Analysis for Missing Data

imputedata function

Data frame with 500 rows and 8 variables:<ul>
<li>MAR_T1: Average value of nonrespondents for continuous outcome in treatment group under MAR assumption.</li>
<li>MAR_C1: Average value of nonrespondents for continuous outcome in control group under MAR assumption.</li>
<li>MAR_T2: Number of success of nonrespondents for binary outcome in treatment group under MAR assumption.</li>
<li>MAR_C2: Number of success of nonrespondents for binary outcome in control group under MAR assumption.</li>
<li>MCAR_T1: Average value of nonrespondents for continuous outcome in treatment group under MCAR assumption.</li>
<li>MCAR_C1: Average value of nonrespondents for continuous outcome in control group under MCAR assumption.</li>
<li>MCAR_T2: Number of success of nonrespondents for binary outcome in treatment group under MCAR assumption.</li>
<li>MCAR_C2: Number of success of nonrespondents for binary outcome in control group under MCAR assumption.</li>
</ul>

imputedata: Imputation results under different methods

Description

Usage

Arguments

Format