Pyramidal Structures: A Preliminary Note
Mario S. Catalani, Giuseppe F. Clerico · Contributions to management science · 1996
We start with the consideration that human decisions are fallible. Let us consider an entrepreneur (if you prefer an owner, a dictator, etc.) who because of his information asymmetry or because of his lack of skills must rely on his collaborators in order to make a decision. To be more specific let the decision consist in accepting or rejecting a project drawn from a project portfolio. Within the portfolio there is a proportion π of good projects. Accepting a good project entails a benefit, while accepting a bad one entails a loss. The entrepreneur has N collaborators, each of whom is identical and independent in the decision making process. Each collaborator has a probability p 1 to accept a good project and a probability p 2 to accept a bad one: hence (1 — p 1 ) is the probability of an error of the first type, and p 2 is the probability of an error of the second type. These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.