Design and analysis of accelerated degradation tests for the stability of biological standards I. Properties of maximum likelihood estimators.

The accelerated degradation test is commonly used to predict the stability of a biological standard during long-term storage at low temperatures. The analysis of complicated by the fact that the standard generally defines its own unit of activity, so only relative rates of degradation at different temperatures can be observed. A series of Monte Carlo simulation studies is described which has been carried out to investigate the accuracy and precision of estimates based on the statistical method and are of use in assessing the extent to which the size and design of the accelerated degradation test influence the precision of the estimate of the low temperature degradation rate.