Validate Input Parameters for Sensitivity Analysis

Usage

validate_all_inputs(
  birth_data_txt = NULL,
  model_prob_csv = NULL,
  conc_data_csv = NULL,
  ndraws = NULL,
  regression_formula = NULL,
  model_type = "mixed",
  model_family = NULL,
  targets = NULL,
  impute_vars = NULL,
  output_dir = NULL,
  mice_m = NULL,
  mice_maxit = NULL,
  mice_method = NULL,
  conc_mean_col = NULL,
  conc_sd_col = NULL,
  conc_raw_col = NULL,
  default_sdlog = 1,
  seed = NULL,
  cat_label = NULL,
  exposure_level_col = "ExposureLevel",
  drop_cat_label_ref = NULL,
  conc_cutoffs = NULL,
  prob_columns = NULL,
  validate_cutoffs = TRUE,
  columns_to_select = NULL,
  rucc_col = NULL,
  id_col = NULL,
  prob_cols = NULL,
  pop_well_col = NULL,
  record_id_col = NULL,
  pwell_col = NULL,
  mice_covs = NULL,
  apply_imputation_fallback = TRUE
)

Arguments

model_prob_csv: Path to modeled probability data CSV file
conc_data_csv: Path to measured concentration data CSV file
ndraws: Number of draws for sensitivity analysis
regression_formula: Regression formula for modeling
model_type: Effects structure of the regression model: one of "mixed", "random", or "fixed"
model_family: Error distribution / link for non-linear (generalized) models, or NULL for a linear model
targets: Target variables for analysis
impute_vars: Variables to impute
output_dir: Output directory for results
mice_m: Number of multiple imputations
mice_maxit: Maximum iterations for MICE
mice_method: MICE method for imputation
conc_mean_col: Column name for concentration lognormal meanlog
conc_sd_col: Column name for concentration lognormal sdlog
seed: Random seed for reproducibility
cat_label: Exposure category label
exposure_level_col: Column name for exposure level
drop_cat_label_ref: Regular expression to drop exposure categories
conc_cutoffs: Concentration cutoffs for exposure categories
prob_columns: Probability columns to include in analysis
validate_cutoffs: Whether to validate cutoffs
columns_to_select: Columns to select from the data
rucc_col: Column name for RUCC classification
id_col: Column name for geographic identifier
prob_cols: Exposure probability columns
pop_well_col: Column name for population well
record_id_col: Column name for birth county
pwell_col: Column name for private well percentage
mice_covs: Covariates for MICE imputation
apply_imputation_fallback: Whether to apply imputation fallback

Value

NULL (throws error if validation fails)