On Nonparametric Predictive Inference for Ordinal Data

Abstract. Nonparametric predictive inference (NPI) is a powerful frequentist statistical framework based only on an exchangeability assumption for future and past observations, made possible by the use of lower and upper probabilities. In this paper, NPI is presented for ordinal data, which are categorical data with an ordering of the categories. The method uses a latent variable representation of the observations and categories on the real line. Lower and upper probabilities for events involving the next observation are presented, and briefly compared to NPI for nonordered categorical data. As an example application the comparison of two groups of ordinal data is presented. Key words: Categorical data; lower and upper probabilities; nonparametric predictive inference; ordinal data; pairwise comparison.

• Type: Conference
• Year: 2010
• Where: IPMU
• Authors: Frank P. A. Coolen, Pauline Coolen-Schrijner, Tahani A. Maturi
• Keywords: Categorical Data, Information Management, IPMU 2010, Nonparametric Predictive Inference, Upper Probabilities