Abstract
The marker-association-segregation-chi 2 (MASC) method with consideration of age, for nonaffected persons, and of age at onset, for affected persons, was applied to a sample of 308 HLA-typed families. Hazard rates modeling the instantaneous risk of catching the disease were estimated under the exponential distribution and with satisfactory goodness of fit. This class of models shows that the hypothesis of the absence of parental imprinting cannot be rejected for insulin-dependent diabetes mellitus.