Sensitivity of probability query

sensquery returns, for a given change in a probability of interest, the parameters' changes to achieve it together with the corresponding CD distances.

Usage

sensquery(
  bnfit,
  interest_node,
  interest_node_value,
  new_value,
  evidence_nodes = NULL,
  evidence_states = NULL
)

Arguments

bnfit: object of class bn.fit.
interest_node: character string. Node of the probability query of interest.
interest_node_value: character string. Level of interest_node.
new_value: numeric value between 0 and 1. New value of the probability of interest.
evidence_nodes: character string. Evidence nodes. Set by default to NULL.
evidence_states: character string. Levels of evidence_nodes. If NULL no evidence is considered. If evidence_nodes="NULL", evidence_states should be set to NULL. Set by default to NULL.

Value

A dataframe with the following columns: node - the vertex of the proposed change; Value node - the level of node to be changed; Value parents - the levels of the parent variables of node; Original value - the original probability defined by Node, Value node and Value parents; Suggested change - the new proposed value for the probability defined by Node, Value node and Value parents; CD distance - the CD distance between the original and new network with the Suggested change.

Details

The Bayesian network should be expressed as a bn.fit object. The name of the node of the probability of interest, its level and the new value should be specified. Evidence could be also indicated. The probability of interest is specified as follows:

P ( interest_node = interest_node_value | evidence_nodes = evidence_states ) = new_value

Only the proportional co-variation scheme is used.

References

Chan, H., & Darwiche, A. (2002). When do numbers really matter?. Journal of artificial intelligence research, 17, 265-287.

Examples