fitBabyMonitor: Fit Baby-MONITOR for CPQCC/VON

In dhelkey/dghrank: Construct Draper-Gittoes-Helkey Institution Rankings

Description

fitBabyMonitor applies the Baby-MONITOR score to a single performance indicator. Designed for CPQCC/VON usage.

Usage

fitBabyMonitor(minimal_data, num_cat, num_cont, subset = FALSE,
  subset_base_catgory = 1, var_intercept = 40, var_cat = 10,
  var_cat_interaction = 10, var_cont = 10, iters = 100, burn_in = 100,
  alpha = 0.01, n_cutoff = 5, bonferroni = TRUE, t_scores = TRUE,
  outcome_na = "set0", subset_na = "category", cat_na = "category",
  cont_na = "median", score_type = "stat_z_scaled", dat_out = FALSE)

Arguments

minimal_data	Data_frame with a particular format: 1st column: Outcome vector (0-1 encoding) 2nd column: Institution ID 3rd column: Subset (if subset == TRUE): Next: num_cat columns of categorical variables (num_cat can equal 0) Next: num_cont columns of continuous variables (num_cont can equal 0)
num_cat	integer. Number of categorical risk adjusters.
num_cont	integer. Number of continious risk adjusters.
subset	logical; if TRUE perform analysis with a subset variable ( and minimal_data must have subset data in the 3rd column)
var_intercept	scalar. Prior variance for the intercept.
var_cat	scalar. Prior variance for categorical risk adjuster parameters.
var_cat_interaction	scalar. Prior variance for interactions between categorical risk adjusters.
var_cont	scalar. Prior variance for continuous risk adjusters.
iters	integer. Desired number of posterior MCMC iterations.
burn_in	integer. Number of 'burn-in' MCMC iterations to discard.
alpha	scalar in (0,1). Statistical signifiance threshold. Posterior (1-alpha)% intervals are generated.
bonferroni	logical; if TRUE, posterior intervals are widened with the Bonferroni correction.
t_scores	logical; if TRUE, posterior intervals confidence intervals constructed with the student t-distribution. If FALSE, z-scores are used.
outcome_na	Method for handling any NA values in the outcome vec. 'remove' removes rows with NA outcomes while 'set0' keeps the row and sets the outcome to 0.
subset_na	Method for handling any NA subset values. 'remove' removes rows with NA subset values while 'category' makes a new subset category (coded as 99) for NA values.
cat_na	Method for handling any NA values in the categorical risk adjusters. 'remove' removes rows with NA values while 'category' makes a new category (coded as 99) for NA catorical risk adjusters.
cont_na	Method for handling any NA values in the continous risk adjusters. 'remove' removes rows with NA values while 'median' replaces NA with the median value of the risk adjuster.
dat_out	logical; if TRUE, export MCMC iterations and other parameters in the dat component.

Value

Returns a large list with the following components:

inst_mat: Matrix (one row per institution) computing summary statistics, DGH institution ranking, and intervals with both effect-size and standardized scores.

full_subset_mat_baseline, full_subset_mat_nobaseline: Matrix with rankings and intervals for the various subset categories.

subset_baseline_mat, subset_nobaseline_mat: Matrix with rankings and intervals for subset categories within institution.

group_labels: A vector of the names of each institution

mcmc_fit: Matrix of MCMC iterations for each coefficient dg_z: Matrix of computed z score for each institution at each MCMC iterations.

coefs: Names of each coefficient (1st is intercept, then institution, then everything else)

prior_var_vec: Vector of prior variances for each coefficient

model_matrix: Design matrix

dhelkey/dghrank documentation built on April 21, 2020, 9:11 a.m.