Bound Generalized Propensity Score

bound_propensity(vals)

Arguments

vals

numeric vector of propensity score estimate values. Note that, for this parameter, the propensity score is (conditional) density and so it ought not be bounded from above.

Value

A numeric vector of the same length as vals, where the returned values are bounded such that the minimum is no lower than 1/n, for the sample size n. @export

Details

Bound estimated values of the generalized propensity score (a conditional density) to avoid numerical instability issues arising from practical violations of the assumption of positivity.