Nondifferentiable errors in beta-compliance integrated logistic models

Abstract

In a classical measurement error model, true regressors. though unknown, vary stochastically around the true level. In inhalation toxicology, environmental epidemiology. and some other problems, the administered dose mav not be completely absorbed into the system, resulting, in a stochastic compliance where the actual dose intake is smaller in magnitude, so the classical measurement error models may not be tenable. We introduce some logistic models tor stochastic compliance of various types incorporating a Btfn compliance distribution The^e compliance-adjusted models are categorized into t hree types: (i) i.ow (or near zero) dose levels, (ii) moderate dose levels, and (iii) high dose levels, resulting in difTeient forms. The buul-T uin method is incorporated to draw statistical conclusions based on BAN estimatois.