Model structure

Each compositional model is represented by its generating sequence - a sequence of probability distributions. Each such a probability distribution is defined over a set of random variables. The sequence of sets of these variables the probability distributions are defined for is call a compositional model structure.

getStructure(model, index = 0)

# S3 method for Model
getVariables(model, index = 0)

Arguments

model

compositional model
index

integer. If not 0, then it returns a vector of variables of respective distribution. It is equivalent to command (getStructure(model))[[index]].

Value

list of vectors of variables

Details

When speaking about a compositional model \(\pi_1(K_1) \triangleright \pi_2(K_2) \triangleright \ldots \triangleright \pi_n(K_n)\), then the sequence \(K_1, K_2, \ldots K_n\) is the structure of the compositional model. The structure can be used to determine some properties of the model - sometimes denoted as structural properties - like *decomposablity*, *conditional independence relations*, etc.

Methods (by class)

  • Model: get Structure of the model

Examples