Bound Generalized Propensity Score

Bound Generalized Propensity Score

Usage

bound_propensity(vals, bound = 0.01)

Arguments

vals

numeric vector of generalized propensity score estimates. Note that the generalized propensity score is a (conditional) density and should only be bounded/truncated away from zero.

bound

numeric atomic giving the lower limit of the generalized propensity score estimates to be tolerated. Estimates less than this will be truncated to this value (default = 0.01). Note that the default may be internally overwritten by reference to the sample size (n), so the lower bound used is the greater of the specified value and 1/n.

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.

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.