Certain frequency distributions which resemble a Poisson distribution, but not quite closely enough, may sometimes be better represented as the sum of two Poissons. A method for the dichotomy and subsequent summation is described, and twenty-nine distributions to which the Poisson was supposedly a good fit, are analyzed.
No takes yet. Share an insight, caveat, or question.
Walter Schilling (1947) studied this question.
Synapse has enriched 5 related papers. Consider them for comparative context: