Nonparametric estimation of effective doses at quantal response

Abstract

For the quantal response model we propose a new direct method for nonparametric estimation of the effective dose level $ED_{100\lambda}$ ($0 < \lambda < 1$). This method yields a simple and reliable monotone estimate of the effective dose level curve $\lambda\to ED_{100\lambda}$ and is appealing to users of conventional smoothing methods of kernel estimates. Moreover, it is computationally very efficient, because it does not require a numerical inversion of the estimate of the quantile dose response curve. We prove asymptotic normality of this new estimator and compare it with the DNP-estimator.