prop2weight: Convert propensity to weight

In awmercer/bestimate: Bayesian Finite Population Survey Estimation With FPBB and BART

Description

Takes a vector of propensity scores for a sample and converts them into propensity weights using their odds. By default, it assumes that the propensities represent the probability than an observation comes from the sample rather than the reference dataset. If the propensity is the probability of being from the reference dataset, set p_denom = FALSE.

Usage

prop2weight(p, p_denom = TRUE)

Arguments

p	A vector of propensity scores
p_denom	If TRUE, indicates that the weights should be calculated as wt = (1 - p)/p. Otherwise wt = p/(1 - p)

Value

A vector of propensity weights

awmercer/bestimate documentation built on May 22, 2019, 8:50 p.m.