feasible_point_search: Method for finding initial points of the EAM algorithm

In depCensoring: Statistical Methods for Survival Data with Dependent Censoring

Method for finding initial points of the EAM algorithm

Description

Also called the 'initialization' step in KMS19, this method tries to find at least one initial feasible point, which is required to run the EAM algorithm. ToDo: Investigate whether the feasible point search of Bei (2024) is better. If so, implement it.

Usage

feasible_point_search(
  test.fun,
  hyperparams,
  verbose,
  picturose = FALSE,
  parallel = FALSE
)

Arguments

test.fun	Function that takes a parameter vector as a first argument and returns the test statistic, as well as the critical value.
hyperparams	List of hyperparameters.
verbose	Verbosity parameter.
picturose	Picturosity flag. If TRUE, a plot illustrating the workings of the algorithm will updated during runtime. Default is picturose = FALSE.
parallel	Flag for whether or not parallel computing should be used. Default is parallel = FALSE.

Value

Results of the initial feasible point search.

References

Kaido et al. (2019). Confidence intervals for projections of partially identified parameters. Econometrica. 87(4):1397-1432.

depCensoring documentation built on April 4, 2025, 1:52 a.m.