# Rh/Fragment 149 22

## < Rh

32.555Seiten in
diesem Wiki

 Typus Verschleierung Bearbeiter Hindemith Gesichtet
Untersuchte Arbeit:
Seite: 149, Zeilen: 22-26
Quelle: Neal 2000
Seite(n): 261, 262, Zeilen: 28-29, 1-4
Following, a Gibbs sampling update is performed for cl in this representation of the posterior distribution. Since ci must be either one of the components associated with other observations or one of the auxiliary components that were introduced, we can easily do Gibbs sampling by evaluating the relative probabilities of these possibilities. Once a new value for cl has been chosen, we discard all values that are not associated with an observation. We now perform a Gibbs sampling update for ci in this representation of the posterior distribution. Since ci must be either one of the components associated with other

[Page 262]

observations or one of the auxiliary components that were introduced, we can easily do Gibbs sampling by evaluating the relative probabilities of these possibilities. Once a new value for ci has been chosen, we discard all ϕ values that are not now associated with an observation.

 Anmerkungen Dem Leser ist zwar klar, dass der Verfasser hier die Anwendung des Algorithmus 8 von Neal auf das untersuchte Problem beschreibt. Nicht klar ist allerdings, dass er dies mit den Worten Neals tut. Sichter (Hindemith), KnallErbse