____ _ _ _ _
| _ \ ___ | |_ (_) _ __ ___ __| | (_) __ _
| |_) | / _ \ | __| | | | '_ \ / _ \ / _| | | | / _ |
| _ < | __/ | |_ | | | |_) | | __/ | (_| | | | | (_| |
|_| \_\ \___| \__| |_| | .__/ \___| \__,_| |_| \__,_|
|_|
- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b- `b
Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―Β―
Posterior predictive distribution
ββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ
top
In der Bayesschen Statistik ist die Posterior predictive distribution eines statistischen Modellscite-ref-1[1] die bedingte Wahrscheinlichkeitsdichte neuer, unbeobachteter Werte x ~ ~ {\displaystyle {\tilde {x}}} , gegeben alle bisherigen Beobachtungen x {\displaystyle \mathbf {x} } . Man erhΓ€lt sie durch Parameter-Integration der bedingten Dichte p ( x ~ ~ | ΞΈ ΞΈ ) {\displaystyle p({\tilde {x}}|\theta )} mit der Posterior-Dichte p ( ΞΈ ΞΈ | x ) {\displaystyle p(\theta |\mathbf {x} )} .
Contents
β’ Definition
β’ Siehe auch
β’ Einzelnachweise
ββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ
Definition
Die Posterior predictive distribution ist definiert als
p ( x ~ ~ | x ) = β« β« Ξ Ξ p ( x ~ ~ | ΞΈ ΞΈ ) p ( ΞΈ ΞΈ | x ) d ΞΈ ΞΈ = E ΞΈ ΞΈ | x [ p ( x ~ ~ | ΞΈ ΞΈ ) ] , {\displaystyle p({\tilde {x}}|\mathbf {x} )=\int _{\Theta }p({\tilde {x}}|\theta )\,p(\theta |\mathbf {x} )\operatorname {d} \!\theta =\mathbb {E} _{\theta |\mathbf {x} }[p({\tilde {x}}|\theta )],}
wobei Ξ Ξ {\displaystyle \Theta } der Parameterraum und p ( ΞΈ ΞΈ | x ) {\displaystyle p(\theta |\mathbf {x} )} die Posterior-Dichte ist. Die Gleichheit lΓ€sst sich mit dem Gesetz der totalen Wahrscheinlichkeit direkt sehen.
Die Posterior predictive distribution spielt zum Beispiel im Rahmen der GauΓ-Prozess-Regression eine wichtige Rolle.
Abgrenzung gegenΓΌber der Prior predictive distribution
Die Prior predictive distribution lΓ€sst die beobachteten Daten auΓer Acht: p ( x ~ ~ ) = β« β« Ξ Ξ p ( x ~ ~ | ΞΈ ΞΈ ) p ( ΞΈ ΞΈ ) d ΞΈ ΞΈ {\displaystyle p({\tilde {x}})=\int _{\Theta }p({\tilde {x}}|\theta )\,p(\theta )\operatorname {d} \!\theta }
Bootstrap predictive distribution
Die Posterior predictive distribution kann durch Anwendung der Bootstrap predictive distribution p B ( x Β― Β― β£ β£ X ) = β« β« p ( x Β― Β― β£ β£ ΞΈ ΞΈ M L E ( X ~ ~ ) ) p ^ ^ ( X ~ ~ ) d X ~ ~ {\displaystyle p_{B}({\bar {x}}\mid X)=\int p({\bar {x}}\mid \theta _{MLE}({\tilde {X}})){\hat {p}}({\tilde {X}})d{\tilde {X}}} genΓ€hert werden, wobei X ~ ~ {\displaystyle {\tilde {X}}} per Bootstrapping-Verfahren aus der empirischen Verteilungsfunktion gezogene Stichproben sind.cite-ref-2[2]cite-ref-3[3]
Siehe auch
Einzelnachweise