This paper presents an optimal strategy for sequential sampling from binomial distributions. The strategy presented is general in that it is a "multi-action" rather than a twoaction procedure. While the major task is to estimate the proportion, p, of "successes" in a hypothetical, infinite population of binary observations, it is assumed that the decision maker is only concerned with which of a...

Power-law distributions have a wide range of applications including physics, economics, and biology. We derive a new family of densities with support on the real line which obeys a broken power law called the volcano distribution. An estimation procedure is also outlined. The volcano density is very flexible as it can be unbounded or bimodal. It allows for an infinite mean or an undefined mean....