R/gof_map.R

In modelsummary: Summary Tables and Plots for Statistical Models and Data: Beautiful, Customizable, and Publication-Ready

#' internal function to build gof_map
#'
#' @noRd
gof_map_build <- function() {
  text <- 'raw, clean, fmt, omit,
  nobs,                      Num.Obs.,           0, FALSE,
  nimp,                      Num.Imp.,           0, FALSE,
  nclusters,                 Num.Clust.,         0, FALSE,
  nblocks,                   Num.Blocks,         0, FALSE,
  r.squared,                 R2,                 3, FALSE,
  r2,                        R2,                 3, FALSE,
  adj.r.squared,             R2 Adj.,            3, FALSE,
  r2.adjusted,               R2 Adj.,            3, FALSE,
  r2.marginal,               R2 Marg.,           3, FALSE,
  r2.conditional,            R2 Cond.,           3, FALSE,
  r2.within,                 R2 Within,          3, FALSE,
  r2.within.adjusted,        R2 Within Adj.,     3, FALSE,
  pseudo.r.squared,          R2 Pseudo,          3, FALSE,
  within.r.squared,          R2 Within,          3, FALSE,
  r.squared.within,          R2 Within,          3, FALSE,
  AIC,                       AIC,                1, FALSE,
  aic,                       AIC,                1, FALSE,
  BIC,                       BIC,                1, FALSE,
  bic,                       BIC,                1, FALSE,
  icc,                       ICC,                1, FALSE,
  logLik,                    Log.Lik.,           3, FALSE,
  F,                         F,                  3, FALSE,
  elpd,                      ELPD,               1, FALSE,
  elpd.se,                   ELPD s.e.,          1, FALSE,
  looic,                     LOOIC,              1, FALSE
  looic.se,                  LOOIC s.e.,         1, FALSE
  waic,                      WAIC,               1, FALSE
  rmse,                      RMSE,               2, FALSE,
  statistic.Weak.instrument, Weak IV F-stat,     1, FALSE,
  weak.instruments,          Weak IV F-stat,     1, FALSE,
  statistic.Wu.Hausman,      Wu-Hausman Chi-Sq., 1, FALSE,
  statistic.Sargan,          Sargan J-stat,      1, FALSE,
  nrow,                      Rows,               0, FALSE,
  ncol,                      Columns,            0, FALSE,
  complete.obs,              Num.Obs.,           0, FALSE,
  na.fraction,               Share missing,      2, FALSE,
  std.error.type,            Std.Errors,         3, FALSE,
  se_type,                   Std.Errors,         3, FALSE,
  vcov.type,                 Std.Errors,         0, FALSE,
  aicc,                      AICC,               1, TRUE,
  agfi,                      AGFI,               0, TRUE,
  cfi,                       CFI,                0, TRUE,
  chisq,                     Chi2,               3, TRUE,
  converged,                 Converged,          0, TRUE,
  deviance,                  Deviance,           2, TRUE,
  df,                        DF,                 0, TRUE,
  df.null,                   DF Null,            0, TRUE,
  df.residual,               DF Resid,           0, TRUE,
  estimator,                 Estimator,          0, TRUE,
  log.loss,                  Log Loss,           3, TRUE,
  missing_method,            Missing Method,     0, TRUE,
  nexcluded,                 Num.Excluded,       0, TRUE,
  ngroups,                   Num.Groups,         0, TRUE,
  norig,                     Num.Orig.,          0, TRUE,
  npar,                      Num.Param.,         0, TRUE,
  null.deviance,             Deviance Null,      2, TRUE,
  p.value,                   p,                  3, TRUE,
  p.value.Sargan,            Sargan p,           3, TRUE,
  p.value.Weak.instrument,   Weak IV p,          3, TRUE,
  weak.instruments.p,        Weak IV p,          3, TRUE,
  p.value.Wu.Hausman,        Wu-Hausman p,       3, TRUE,
  pcp,                       PCP,                3, TRUE,
  r2.nagelkerke,             R2 Nagelkerke,      2, TRUE,
  r2.tjur,                   R2 tjur  ,          2, TRUE,
  rmsea,                     RMSE A,             0, TRUE,
  rmsea.conf.high,           RMSE A CI,          0, TRUE,
  score.log,                 Score Log,          3, TRUE,
  score.spherical,           Score Spherical,    3, TRUE,
  sigma,                     Sigma,              3, TRUE,
  srmr,                      SRMR,               0, TRUE,
  statistic,                 Statistics,         3, TRUE,
  wu.hausman,                Wu-Hausman,         3, TRUE,
  wu.hausman.p,              Wu-Hausman (p),     3, TRUE,
  statistic,                 Statistics,         3, TRUE,
  tli,                       TLI,                0, TRUE,'
  out <- utils::read.csv(
    text = text,
    colClasses = c("character", "character", "numeric", "logical", "NULL"))
  for (i in 1:2) {
    out[[i]] <- trimws(out[[i]])
  }
  out
}

#' Data.frame used to clean up and format goodness-of-fit statistics
#'
#' By default, this data frame is passed to the 'gof_map' argument of the
#' 'modelsummary' function. Users can modify this data frame to
#' customize the list of statistics to display and their format. See example
#' below.
#' @docType data
#' @keywords datasets
#' @name gof_map
#' @format data.frame with 4 columns of character data: raw, clean, fmt, omit
#' @examples
#' if (identical(Sys.getenv("pkgdown"),  "true")) {
#'
#' library(modelsummary)
#' mod <- lm(wt ~ drat, data = mtcars)
#' gm <- modelsummary::gof_map
#' gm$omit[gm$raw == 'deviance'] <- FALSE
#' gm$fmt[gm$raw == 'r.squared'] <- "%.5f"
#' modelsummary(mod, gof_map = gm)
#' }
#'
#' @export
gof_map <- gof_map_build()