make_NIW_belief_from_data: Make NIW belief from data.

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

Make NIW belief from data.

Description

Constructs an NIW_belief or NIW_ideal_adaptor object, representing Normal-Inverse Wishart (NIW) parameters for all categories found in the data. This object can be used as a prior for functions like update_NIW_beliefs_incrementally.

Usage

make_NIW_belief_from_data(
  data,
  group = NULL,
  category = "category",
  cues,
  kappa = nu,
  nu = length(cues) + 2,
  verbose = F
)

make_NIW_prior_from_data(
  data,
  group = NULL,
  category = "category",
  cues,
  kappa = nu,
  nu = length(cues) + 2,
  verbose = F
)

make_NIW_ideal_adaptor_from_data(
  data,
  group = NULL,
  category = "category",
  cues,
  kappa = nu,
  nu = length(cues) + 2,
  ...,
  verbose = F
)

Arguments

data	The tibble or data.frame from which to construct the prior.
group	Optionally, one or more grouping variables can be specified. If group is not NULL, one NIW_belief or ideal adaptor will be derived for each level of group. (default: NULL)
category	Name of variable in data that contains the category information. (default: "category")
cues	Name(s) of variables in data that contain the cue information.
kappa	The strength of the beliefs over the category mean (pseudocounts). (default: same as nu)
nu	The strength of the beliefs over the category covariance matrix (pseudocounts). (default: number of cues + 2)
verbose	If true provides more information. (default: FALSE)
prior	Optionally specify a prior probability for each category (in each group). (default: a uniform prior over all categories).
lapse_rate	Optionally specify a lapse rate. (default: NA)
lapse_bias	Optionally specify a lapse bias. (default: NA)
Sigma_noise	Optionally specify a (multivariate Gaussian) covariance matrix of perceptual noise. This argument will be ignored if NULL. (default: NULL)

Details

Currently, make_NIW_prior_from_data() does not infer kappa or nu, nor does it fit hierarchical data. Rather the function simply estimates the category mean and covariance matrix from the sample (data), assumes them to be the expected category mean (mu) and covariance (Sigma), and derives the m and S parameters of the NIW from mu and Sigma based on the user-provided kappa and nu. That means m = mu and S = Sigma * (nu - D -1), where D is the dimensionality of the data.

Value

A tibble that is an NIW_belief or NIW_ideal_adaptor object.

See Also

TBD

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