| get_var_bound {geocausal} | R Documentation |
Calculate variance upper bounds
Description
A function that calculates variance upper bounds
Usage
get_var_bound(estimates, bound_est = 1)
Arguments
estimates |
an object returned from 'get_est()' function |
bound_est |
an integer specifying the estimator for the asymptotic variance bound. Use '1' for the first estimator and '2' for the second estimator.Default is '1'. See details. |
Details
'get_var_bound()' is an internal function to 'get_estimates()' function, performing the estimation analysis after 'get_est()' function. There are two consistent estimators for the asymptotic variance bound. The first estimator (default, 'bound_est=1') performs well in scenarios with low variability, but exhibit some under coverage issue for high variability scenarios where longer time series are needed for getting robust estimator. The second estimator('bound_est=2') yields good coverage rates in simulations, but gives unreasonably large bound when there 'are extremely large IPW weights. Users should carefully consider the characteristics of their data when selecting the estimator.
Value
list of variance upper bounds for each scenario ('bound_haj') and causal contrasts ('bound_tau_haj'). Note that this function returns variance upper bounds for Hajek estimators