Estimation of discrete multivariate densities for computer-aided differential diagnosis of disease
用于计算机辅助疾病鉴别诊断的离散多元密度估计
A Bayesian approach is taken to the problem of estimating a multivariate discrete density. Prior distributions on the space of multivariate densities are formulated to reflect the idea that the density is near a particular one, or near one which has independent variates. Given a sample of multivariate observations, an approximation to the posterior distribution of the unknown density is developed, and simple formulae for a density which approximates the mode of this…