This function is designed for use within weighting() and assess().'
Usage
.make.covariate.table(
data,
sample_indicator,
covariates,
sample_weights = NULL,
estimation_method = "lr",
disjoint_data = TRUE
)Arguments
- data
Dataframe comprised of "stacked" sample and target population data
- sample_indicator
Binary variable denoting sample membership (1 = in sample, 0 = out of sample)
- covariates
Vector of covariates in dataframe that predict sample membership
- sample_weights
Name of column in dataframe holding weights for calculating weighted sample means of covariates in dataframe. If NULL, sample means are unweighted.
- estimation_method
Method to estimate the probability of sample membership. Default is logistic regression ("lr"). Other methods supported are Random Forests ("rf") and Lasso ("lasso").
- disjoint_data
Logical. Defaults to TRUE. If TRUE, then sample and population data are considered disjoint. This affects calculation of the weights.