R/modelSpace.R

In rmsBMA: Reduced Model Space Bayesian Model Averaging

#' Model space for the Trade_data_small dataset
#'
#' A pre-computed model space object obtained using the
#' \code{\link{model_space}} function on the \code{Trade_data_small}
#' dataset with a maximum of 7 regressors and the Unit Information Prior.
#'
#' The object contains all possible linear regression models constructed
#' from the available regressors (up to 7 included simultaneously),
#' together with their estimated coefficients, standard errors,
#' log-marginal likelihood values, R-squared statistics, degrees of
#' freedom, and dilution prior components.
#'
#' The g-prior specification corresponds to the Unit Information Prior
#' (UIP), i.e. \eqn{g = 1/m}, where \eqn{m} denotes the sample size.
#'
#' @format A list of length 5 with the following components:
#' \describe{
#'   \item{x_names}{Character vector containing the names of the regressors.}
#'   \item{ols_results}{Matrix containing the full model space. Each row
#'   corresponds to a model specification and includes:
#'   \itemize{
#'     \item binary inclusion indicators for regressors,
#'     \item estimated coefficients (including intercept),
#'     \item standard errors,
#'     \item log-marginal likelihood,
#'     \item \eqn{R^2},
#'     \item degrees of freedom,
#'     \item dilution prior term.
#'   }}
#'   \item{MS}{Integer. Total number of models in the model space.}
#'   \item{M}{Integer. Maximum number of regressors allowed in each model.}
#'   \item{K}{Integer. Total number of available regressors.}
#' }
#'
#' @source Generated using:
#' \preformatted{
#' modelSpace <- model_space(Trade_data_small, M = 7, g = "UIP")
#' }
#'
#' @seealso \code{\link{model_space}}
#'
#' @keywords datasets
#' @name modelSpace
NULL