controlInf
constructs a list with all necessary control parameters
for statistical inference.
Usage
controlInf(
vars_selection = FALSE,
var_method = c("analytic", "bootstrap"),
rep_type = c("auto", "JK1", "JKn", "BRR", "bootstrap", "subbootstrap", "mrbbootstrap",
"Fay"),
bias_inf = c("union", "div"),
num_boot = 500,
bias_correction = FALSE,
alpha = 0.05,
cores = 1,
keep_boot,
pmm_exact_se = FALSE,
pi_ij
)
Arguments
- vars_selection
If
TRUE
, then variables selection model is used.- var_method
variance method.
- rep_type
replication type for weights in the bootstrap method for variance estimation passed to
survey::as.svrepdesign()
. Default issubbootstrap
.- bias_inf
inference method in the bias minimization.
if
union
then final model is fitting on union of selected variables for selection and outcome modelsif
div
then final model is fitting separately on division of selected variables into relevant ones for selection and outcome model.
- num_boot
number of iteration for bootstrap algorithms.
- bias_correction
if
TRUE
, then bias minimization estimation used during fitting the model.- alpha
Significance level, Default is 0.05.
- cores
Number of cores in parallel computing.
- keep_boot
Logical indicating whether statistics from bootstrap should be kept. By default set to
TRUE
- pmm_exact_se
Logical value indicating whether to compute the exact standard error estimate for
pmm
estimator. The variance estimator for estimation based onpmm
can be decomposed into three parts, with the third being computed using covariance between imputed values for units in probability sample using predictive matches from non-probability sample. In most situations this term is negligible and is very computationally expensive so by default this is set toFALSE
, but it is recommended to set this value toTRUE
before submitting final results.- pi_ij
TODO, either matrix or
ppsmat
class object.
See also
nonprob()
-- for fitting procedure with non-probability samples.