scientific article; zbMATH DE number 3888710

From MaRDI portal
Publication:3221194





[For the entire collection see Zbl 0517.00012.] For a countable mixture \(f(x)=\sum p_ ih(x| \mu_ i,\sigma_ i)\) of normal distributions h a prior distribution for \(p_ i,\mu_ i,\sigma_ i\) is used with \((p_ 1,p_ 2,.)\) and \((\mu_ 1,\mu_ 2,...,\sigma_ 1,\sigma_ 2,.)\) independent, \((\mu_ 1,\sigma_ 1)\), \((\mu_ 2,\sigma_ 2)..\). i.i.d. gamma normal and \(p_ j=\prod^{j- 1}_{i=1}(1-q_ j)\) with \(q_ 1,q_ 2,..\). i.i.d. beta. The prior can be described as a Dirichlet process. \(E(f(x)| x_ 1,...,x_ n)\) is computed for a sample \(X_ 1,...,X_ n.\) Interpretation of the prior distribution (how much do you trust your prior guess a.o.) and choice of the prior parameters are discussed. Computations are made using a Monte Carlo technique of L. Kuo.




Cited in
(only showing first 100 items - show all)








This page was built for publication:

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3221194)