posterior.CatDP {bbricks} | R Documentation |
For the model structure:
pi|alpha \sim DP(alpha,U)
x|pi \sim Categorical(pi)
where DP(alpha,U) is a Dirichlet Process on positive integers, alpha is the "concentration parameter" of the Dirichlet Process, U is the "base measure" of this Dirichlet process, it is an uniform distribution on all positive integers.Categorical() is the Categorical distribution. See dCategorical
for the definition of the Categorical distribution.
In the case of CatDP, x can only be positive integers.
Update prior knowledge by adding the information of newly observed samples x. The model structure and prior parameters are stored in a "CatDP" object, the prior parameters in this object will be updated after running this function.
## S3 method for class 'CatDP' posterior(obj, ss, w = NULL, ...)
obj |
A "CatDP" object. |
ss |
Sufficient statistics of x. In Categorical-DP case the sufficient statistic of sample x can either be an object of type "ssCatDP" generated by sufficientStatistics(), or x itself(if x is a integer vector with all positive values). |
w |
Sample weights, default NULL. |
... |
Additional arguments to be passed to other inherited types. |
None. the model stored in "obj" will be updated based on "ss".
Teh, Yee W., et al. "Sharing clusters among related groups: Hierarchical Dirichlet processes." Advances in neural information processing systems. 2005.
CatDP
,posteriorDiscard.CatDP
,sufficientStatistics.CatDP
## generate some integer samples x <- sample(1L:10L,size = 40,replace = TRUE) obj <- CatDP() obj2 <- CatDP() obj3 <- CatDP() ## update CatDP object with sufficient statistics ss <- sufficientStatistics(obj=obj,x=x) posterior(obj = obj,ss = ss) ## or, update with x itself posterior(obj = obj2,ss = x) ## or, update with x itself, one by one for(xx in x) posterior(obj = obj3,ss = xx) ## obj, obj2, obj3 should be the same: obj obj2 obj3