*Since version 7.18*

Impact analysis evaluates the effect of different subsets of evidence on a target variable and/or state. The target is known as the **Hypothesis**.

Impact analysis can be used to:

- See which individual pieces of evidence change the target the most
- See which individual pieces of evidence, when excluded, change the target the most
- See which subsets of evidence which, when included or excluded, change the target the most.

Variable type | Hypothesis | Evidence |
---|---|---|

Discrete | Yes | Yes |

Continuous | Yes | Yes |

Hybrid | Yes | Yes |

We use the term

Noneto refer to the case when none of the evidence (being analyzed) is included.

We use the term

Allto refer to the case when all of the evidence (being analyzed) is included.

The subsets to consider are generated by looking at all possible combinations of evidence, subject to the following constraints.

The number of items of evidence in each subset.

When **Subset method** is set to **Include**, evidence on variables in the subset is included, but not other evidence variables (being analyzed).

When **Subset method** is set to **Exclude**, all evidence is used, except for those in the subset.

For each evidence subset, impact analysis calculates a number of statistics:

This is the KL-Divergence D(P||Q) between the target distribution given the current subset evidence P, and the target distribution given *None* Q.

This tells us how different the target distribution is from the current evidence subset to when no evidence (being analyzed) is set.

This is the KL-Divergence D(P||Q) between the target distribution given *All* P, and the target distribution given this evidence subset Q.

This tells us how different the target distribution is from when all the evidence (being analyzed) is set vs our evidence subset.

When a discrete hypothesis state is included, this reports the probability of that state given the current evidence subset.

When a discrete hypothesis state is included, this reports the difference between the probability of that state given the current evidence subset and no evidence (being analyzed).

When a discrete hypothesis state is included, this reports the difference between the probability of that state given *All* the evidence and the current evidence subset.

When a discrete hypothesis state is included, this reports the ratio between the probability of that state given the current evidence subset and no evidence (being analyzed).

This is also known as the

Normalized Likelihoodin impact analysis.

When a discrete hypothesis state is included, this reports the ratio between the probability of that state given *All* the evidence and the current evidence subset.