Adding new factors to predictive models : the effect on the risk of extrapolation

摘要: Abstract Two aspects of the addition new environmental factors to predictive microbiology models are discussed. The concept risk extrapolation is introduced characterize probability that a prediction outside model interpolation region. For an empirical model, region defined by data which used estimate parameters model. It shown can become unexpectedly high during extending describe effect newer factors, if extension supported insufficient data. A convenient method also presented accommodate observations on additional factors. New be added basic common in such way will special case extended models. Conditions under this approach useful Published illustrate both points.