aggregate_models_by_group_structure: Aggregate multiple ideal observers/adaptors into an average...

In hlplab/MVBeliefUpdatr: Fitting, Summarizing, and Visualizing of Multivariate Gaussian Ideal Observers and Adaptors

Aggregate multiple ideal observers/adaptors into an average ('typical') ideal observer/adaptor

Description

Takes a grouped collection of ideal observers or adaptors (or MVG / NIW_belief objects) and averages them into a single object. Aggregation proceeds sequentially from the first to the last element of group_structure: first, a mean and covariance matrix will be obtained for all combinations of grouping variables; then the first grouping variable is collapsed over, creating an average for all combinations of the remaining grouping variables; and so on. For this process to have a reasonable outcome, it is important that the grouping variables are hierarchically organized from more specific to less specific: the second group variable should be a superset of the first group variable; the third group variable should be a superset of the second group variable; etc.

Usage

aggregate_models_by_group_structure(x, group_structure = NULL)

Arguments

x	An MVG, MVG_ideal_observer, NIW_belief, or NIW_ideal_adaptor object.
group_structure	The group structure that will be used for aggregation.

Value

The aggregated object.

See Also

TBD

hlplab/MVBeliefUpdatr documentation built on March 29, 2025, 10:42 p.m.