Dose-response models are intensively used in herbicide bioassays. Despite r
ecent advancements in the development of new herbicides, statistical analys
es are commonly based on asymptotic approximations that are sometimes poor.
This paper presents the use of recent results in higher order asymptotics
for Likelihood-based inference in nonlinear regression. The methods present
ed provide accurate approximation for the distribution of test statistics a
nd for prediction limits. Analyses of the fit and measures of detection lim
its of the bioassays are considered, and the potential of the methods is il
lustrated by examples with real data.