ON FITTING A POPULATION-MODEL IN THE PRESENCE OF OBSERVATION ERROR

Observation error can lead to substantial bias in estimating the param eters of a population model. While maximum-likelihood estimation is po ssible in principle, it can be extremely difficult in practice due to the complicated behavior of the likelihood function. This note describ es a simple method that can be used to fit a population model in the p resence of observation error. The method is illustrated using the disc rete-time logistic model.