bootstrap_se {CIEE}

Bootstrap standard error estimates

Description

Function to obtain bootstrap standard error estimates for the parameter estimates of the get_estimates function, under the generalized linear model (GLM) or accelerated failure time (AFT) setting for the analysis of a normally-distributed or censored time-to-event primary outcome.

Usage

bootstrap_se(setting = "GLM", BS_rep = 1000, Y = NULL, X = NULL,
  K = NULL, L = NULL, C = NULL)

Arguments

setting

String with value "GLM" or "AFT" indicating whether standard error estimates are obtained for a normally-distributed ("GLM") or censored time-to-event ("AFT") primary outcome Y.

BS_rep

Integer indicating the number of bootstrap samples that are drawn.

Y

Numeric input vector for the primary outcome.

X

Numeric input vector for the exposure variable.

K

Numeric input vector for the intermediate outcome.

L

Numeric input vector for the observed confounding factor.

C

Numeric input vector for the censoring indicator under the AFT setting (must be coded 0 = censored, 1 = uncensored).

Details

Under the GLM setting for the analysis of a normally-distributed primary outcome Y, bootstrap standard error estimates are obtained for the estimates of the parameters \alpha_0, \alpha_1, \alpha_2, \alpha_3, \sigma_1^2, \alpha_4, \alpha_{XY}, \sigma_2^2 in the models

Y = \alpha_0 + \alpha_1 \cdot K + \alpha_2 \cdot X + \alpha_3 \cdot L + \epsilon_1, \epsilon_1 \sim N(0,\sigma_1^2)

Y^* = Y - \overline{Y} - \alpha_1 \cdot (K-\overline{K})

Y^* = \alpha_0 + \alpha_{XY} \cdot X + \epsilon_2, \epsilon_2 \sim N(0,\sigma_2^2),

accounting for the additional variability from the 2-stage approach.

Under the AFT setting for the analysis of a censored time-to-event primary outcome, bootstrap standard error estimates are similarly obtained of the parameter estimates of \alpha_0, \alpha_1, \alpha_2, \alpha_3, \sigma_1, \alpha_4, \alpha_{XY}, \sigma_2^2

Value

Returns a vector with the bootstrap standard error estimates of the parameter estimates.

Examples

dat <- generate_data(setting = "GLM", n = 100)

# For illustration use here only 100 bootstrap samples, recommended is using 1000
bootstrap_se(setting = "GLM", BS_rep = 100, Y = dat$Y, X = dat$X,
             K = dat$K, L = dat$L)