On an asymptotic comparison of maximum likelihood and berkson's modified minimum chi-square estimates in the two parameter dose response models

Abstract

This paper focuses on three standard dose response models, namely, the multistage Weibull, logistic and log-logistic models, each involving two unknown parameters. The maximum likelihood and Berkson's modified minimum chi-square methods are employed to estimate the two unknown parameters in each model, and the mean squared errors of these estimates are derived asymptotically to the order of approximation of n- 2 , where n is the sample size.The AIC and deviance criteria are used for model selection. One real data set is used for illustration. The results show that, as expected, the mean squared errors of the estimates behave differently for different dose groups and dose levels. © Nova Science Publishers, Inc.