The influence of design parameters on statistical inference in
non-linear estimation; a simulation study based on survival data and
hazard modelling
Andersen, J. S., Bedaux, J. J. M., Kooijman, S. A. L. M. and
Holst, H. 2000. The influence of design parameters on statistical
inference in non-linear estimation; a simulation study based on
survival data and hazard modelling. Journal of Agricultural,
Biological and Environmental Statistics, 5: 323 - 341.
Abstract
This paper descibes the influence of design characteristics on the
statistical inference for an ecotoxicological hazard-based model,
using simulated survival data. The design characteristics of interest
are the number and spacing of observations (counts) in time, the
number and spacing of exposure concentrations (within cmin
and cmax) and the initial number of individuals at time 0
in each concentration. A comparison of the coverage probabilities for
confidence limits arising from the profile-likelihood approach and the
Wald-based approach is carried out. The Wald-based approach is very
sensitive to the choice of design characteristics whereas the
profile-likelihood approach is more robust and unbiased. Special
attention is paid to estimating a parametric no-effect concentration
in realistic small-sample situations, since this is the most
interesting parameter from an environmental protection point of view.