Enhanced Tipping Point Displays the Results of Sensitivity
Analyses for Missing Data
Using the idea of "tipping point" (proposed in
Gregory Campbell, Gene Pennello and Lilly Yue(2011)
<10.1080>) 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 the values of missing data in two treatment arms, calculate
corresponding estimated treatment effects and p values and draw
a colored heat-map to visualize them. 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.10.1080>
- A function created based on Rcpp is not essential currently, so we plan to remove it to make it more convenient for users to install the pkg without compilation.
Enhanced Tipping Point Displays the Results of Sensitivity Analyses for Missing Data.