test.point_Bei_MT: Perform the test of Bei (2024) simultaneously for multiple...

In depCensoring: Statistical Methods for Survival Data with Dependent Censoring

Perform the test of Bei (2024) simultaneously for multiple time points.

Description

This function performs the unconditional moment restriction test as described in Bei (2024). This function directly extends test.point_Bei by allowing for pairs of moment restrictions over a grid of time points.

Usage

test.point_Bei_MT(
  r,
  c,
  t,
  par.space,
  data,
  hp,
  verbose = FALSE,
  inst.func.evals = NULL,
  alpha = 0.95,
  parallel = FALSE
)

Arguments

r	Result of the projection for which the test should be carried out.
c	The projection matrix. For now, c is restricted to being an elementary vector, i.e. c = (0, ...,0, 1, 0, ..., 0).
t	The time point at which to evaluate theta. Also allowed to be a vector of time points (used in estimating the model under assumed time- independent coefficients).
par.space	Matrix containing 2 columns and d_\theta rows, where d_\theta is the dimension of the parameter space. The first column represents the lower left corner of the parameter space, the second column represents the upper right corner. At least for the time being, only rectangular parameter spaces are allowed.
data	Data frame on which to base the test.
hp	List of hyperparameters needed.
verbose	Boolean variable indicating whether to print updates of the estimation process to the console.
inst.func.evals	Matrix of precomputed instrumental function evaluations for each observation in the data set. Used to speed up the simulations. If NULL, the evaluations will be computed during execution of this function. Default is inst.func.evals = NULL.
alpha	The significance level at which to perform the test. Default is alpha = 0.95.
parallel	Flag for whether or not parallel computing should be used. Default is parallel = FALSE.

References

Bei, X. (2024). Local linearization based subvector inference in moment inequality models. Journal of Econometrics, 238(1), 105549-. https://doi.org/10.1016/j.jeconom.2023.10554

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