Minor documentation updates following JOSS publication.
Release corresponding to JOSS paper.
Adds automated testing for Step 1 wrapper functions (e.g., vt1_lasso) and
get_mnpp.
Adds support for classification trees in Step 2 by setting
step2 = 'classtree' with a given threshold of threshold.
Adds the print.tunevt method.
Fixes a bug where zbar was calculated using the mean difference in the first
column of the data instead of using the location of the variable Y.
Adds the parallel option to tunevt to support parallel backends.
This patch reconciles an invalid URI in the tunevt documentation's references.
This is a new package that implements the Virtual Twins algorithm for subgroup identification (Foster et al., 2011) while controlling the probability of falsely detecting differential treatment effects when the conditional treatment effect is constant across the population of interest. These methods were originally presented in Wolf et al. (2022).
Foster, J. C., Taylor, J. M., & Ruberg, S. J. (2011). Subgroup identification from randomized clinical trial data. Statistics in Medicine, 30(24), 2867–2880. https://doi.org/10.1002/sim.4322
Wolf, J. M., Koopmeiners, J. S., & Vock, D. M. (2022). A permutation procedure to detect heterogeneous treatment effects in randomized clinical trials while controlling the type-I error rate. Clinical Trials. https://doi.org/10.1177/17407745221095855
tunevt() fits a Virtual Twins model using user-specified Step 1 and Step 2
models with parameter selection to control the probability of a false
discovery.