BACKGROUND We introduce a new criterion, the percentile inclusion probability, for comparing methods for calculating reference intervals. The criterion is compared with a previously published measure of reliability suggested by Linnet (Linnet K. Clin Chem 1987;33:381-6), the ratio of the width of the confidence interval for the percentile to that of the reference interval. METHODS Data were s...

Let X=(X1 ,X2 ) be a continuous random vector. Under the assumption that the marginal distributions of X1 and X2 are given, we develop models for vector X when there is partial information about the dependence structure between X1  and X2. The models which are obtained based on well-known Principle of Maximum Entropy are called the maximum entropy (ME) mo...

Cluster analysis is reformulated as a problem of estimating the para- meters of a mixture of multivariate distributions. The maximum-likelihood theory and numerical solution techniques are developed for a fairly general class of distributions. The theory is applied to mixtures of multivariate nor- mals (NORMIX) and mixtures of multivariate Bernoulli distributions (Latent Classes). The feasibili...

“Mixel” is a heterogeneous pixel which consists of more than two classification classes. This paper discusses the statistical properties of mixels, and also develops the theory of mixture density estimation in the presence of mixels. First we show the essential importance of the concept of stable distributions with regard to the analysis of mixels; next we derive a new statistical model called ...

Generalized linear models with scale mixture of multivariate normal (SMMVN) link functions are considered to model correlated ordinal response data. A unified Bayesian approach, which includes prior elicitation and model comparison, is proposed. In order to incorporate available prior information, we propose a class of informative prior distributions on model parameters. The propriety of the pr...

We consider maximum likelihood estimation of finite mixture of uniform distributions. We prove that the maximum likelihood estimator is strongly consistent, if the scale parameters of the component uniform distributions are restricted from below by exp(−nd), 0 < d < 1, where n is the sample size.