Summary A mixture from exponential family is compared with a natural approximation in the family. It is shown that the two resulting densities must cross each other exactly twice and in a prescribed manner. A deeper understanding of the behaviour of some well known sets of data is obtained. Applications in reliability theory and for symmetric random walks are described.
No takes yet. Share an insight, caveat, or question.
Moshe Shared (1980) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: