autoMrP: Improving MrP with Ensemble Learning

#' Quasi census data.
#'
#' The census file is generated from the full 2008 National Annenberg Election
#' Studies item CBb01 by dissaggregating the 64 ideal type combinations of the
#' individual level variables L1x1, L2x2 and L1x3. A row is an ideal type in a
#' given state.
#'
#'
#' @format A data frame with 2934 rows and 13 variables:
#' \describe{
#'   \item{state}{U.S. state}
#'   \item{L2.unit}{U.S. state id}
#'   \item{region}{U.S. region (four categories: 1 = Northeast; 2 = Midwest; 3 = South; 4 = West)}
#'   \item{L1x1}{Age group (four categories)}
#'   \item{L1x2}{Education level (four categories)}
#'   \item{L1x3}{Gender-race combination (six categories)}
#'   \item{freq}{State-level frequency of ideal type}
#'   \item{proportion}{State-level proportion of respondents of that ideal type in the population}
#'   \item{L2.x1}{State-level share of votes for the Republican candidate in the previous presidential election}
#'   \item{L2.x2}{State-level percentage of Evangelical Protestant or Mormon respondents}
#'   \item{L2.x3}{State-level percentage of the population living in urban areas}
#'   \item{L2.x4}{State-level unemployment rate}
#'   \item{L2.x5}{State-level share of Hispanics}
#'   \item{L2.x6}{State-level share of Whites}
#' }
#' @usage data(taxes_census)
#' @source The data set (excluding L2.x3, L2.x4, L2.x5, L2.x6) is taken from the
#'   article: Buttice, Matthew K, and Benjamin Highton. 2013. "How does
#'   multilevel regression and poststrat-stratification perform with
#'   conventional national surveys?" Political Analysis 21(4): 449-467. L2.x3,
#'   L2.x3, L2.x4, L2.x5 and L2.x6 are available at
#'   \url{https://www.census.gov}.
"taxes_census"