se.coef.dr {mable}

Standard errors of coefficients in density ratio model

Description

Bootstrap estimates of standard errors for the regression coefficients which are estimated by maximum approximate Bernstein/Beta likelihood estimation method in a density ratio model based on two-sample raw data.

Usage

se.coef.dr(
  obj,
  grouped = FALSE,
  B = 500L,
  parallel = FALSE,
  ncore = NULL,
  controls = mable.ctrl()
)

Arguments

obj

Class 'mable_dr' object return by mable.dr or mable.dr.group functions

grouped

logical: are data grouped or not.

B

number of bootstrap runs.

parallel

logical: do parallel or not.

ncore

number of cores used for parallel computing. Default is half of availables.

controls

Object of class mable.ctrl() specifying iteration limit and the convergence criterion for EM and Newton iterations. Default is mable.ctrl. See Details.

Details

Bootstrap method is used based on bootstrap samples generated from the MABLE's of the densities f0 and f1. The bootstrap samples are fitted by the Bernstein polynomial model and the glm() to obtain bootstrap versions of coefficient estimates.

Value

the estimated standard errors