kw.wt: Calculate KW pseudo-weights given propensity scores
In chkern/KWML: Boosted Kernel Weighting

View source: R/weighting_functions.R

kw.wt

R Documentation

Calculate KW pseudo-weights given propensity scores

Description

This function computes KW pseudo-weights based on propensity scores that are provided by the user.

Usage

kw.wt(
  p_score.c,
  p_score.s,
  svy.wt,
  h = NULL,
  mtch_v = NULL,
  krn = "triang",
  large = F,
  rm.s = F
)

Arguments

`p_score.c`	Predicted propensity scores for the non-probability sample
`p_score.s`	Predicted propensity scores for the probability sample
`svy.wt`	A vector of survey weights for the probability sample units
`h`	Bandwidth parameter (will be calculated corresponding to kernel function if not specified)
`krn`	Kernel function. "`triang`": triangular density on (-3, 3), "`dnorm`": standard normal density, "`dnorm_t`": truncated standard normal density on (-3, 3).
`large`	The cohort size is so large that it has to be divided into pieces. Default is `FALSE`.
`rm.s`	Remove unmatched survey units or not. Default is `FALSE`.

Details

If there are unmatched survey sample units, the program gives "The input bandwidth h is too small. Please choose a larger one!" If rm.s=TRUE, the program deletes unmatched survey sample units, and gives a warning "records in the prob sample were not used because of a small bandwidth" If rm.s=FALSE, the program evenly distributes weights of unmatched survey sample units to all non-probability sample units.