Keywords: Bayesian inference, Bootstrapping, Constraint optimization, Maximum likelihood estimation, Risk difference.
For estimating adjusted risk differences, we propose several constraint approaches based on binomial and Poisson regression models. The proposed models span the areas of ordinary least squares, maximum likelihood estimation and Bayesian inference. Comparing to existing approaches, our methods prevent estimates and confidence intervals of predicted probabilities from falling out of the valid range. Through extensive simulation studies, we demonstrate that the proposed methods solve the issue of having estimates or confidence limits of predicted probabilities out of (0, 1), while offering performance comparable to its alternative in terms of the bias, variability and coverage rates in point and interval estimation of the intercept and RD. An application study is performed using data from Prospective Registry Evaluating Myocardial Infarction: Event and Recovery (PREMIER) study.