Statistics Based on Dirichlet Processes and Related Topics

This book focuses on the properties associated with the Dirichlet process, describing itsusea priorifor nonparametric inference and the Bayes estimateto obtain limitsfor the estimable parameter.It presentsthe limits and the well-knownU- andV-statistics as a convex combination ofU-statistics, and by investigating this convex combination, it demonstratesthese three statistics. Next, the book notes that the Dirichlet process gives the discrete distribution with probability one, even if the parameter of the process is continuous. Therefore, there are duplications among the sample from the distribution, whicharediscussed. Because sampling from the Dirichlet process is described sequentially, it can be described equivalently by the Chinese restaurant process. Using this process, the Donnelly-Tavaré-Griffiths formulas I and II are obtained, both of which give the Ewens' samplingformula. The book then shows the convergence and approximation of the distribution for its number of distinct components.Lastly, it explains the interesting propertiesof the Griffiths-Engen-McCloskey distribution,which is related to theDirichlet process and the Ewens' sampling formula.