Perform covariate selection for linear ER model
dev_ermod_bin_cov_sel.RdThis functions is used to develop an ER model with covariates for binary
and continuous endpoints.
projpred package is used for variable selection.
Usage
dev_ermod_bin_cov_sel(
data,
var_resp,
var_exposure,
var_cov_candidates,
cv_method = c("LOO", "kfold"),
k = 5,
validate_search = FALSE,
nterms_max = NULL,
.reduce_obj_size = TRUE,
verbosity_level = 1,
chains = 4,
iter = 2000
)
dev_ermod_lin_cov_sel(
data,
var_resp,
var_exposure,
var_cov_candidates,
cv_method = c("LOO", "kfold"),
k = 5,
validate_search = FALSE,
nterms_max = NULL,
.reduce_obj_size = TRUE,
verbosity_level = 1,
chains = 4,
iter = 2000
)Arguments
- data
Input data for E-R analysis
- var_resp
Response variable name in character
- var_exposure
Exposure variable names in character
- var_cov_candidates
Candidate covariate names in character vector
- cv_method
Cross-validation method. Default is "LOO" (recommended). Use "kfold" if you see warnings on Pareto k estimates.
- k
Number of folds for kfold CV. Only used if cv_method is "kfold".
- validate_search
Whether to validate the search. Default is FALSE. Recommend to set to TRUE for kfold CV. Do not use for LOO (run time would become too long).
- nterms_max
Maximum number of terms to consider in the model. Default is NULL (all terms are considered).
- .reduce_obj_size
Whether to reduce object size by removing some elements from projpred outputs that are not necessary for the functionality of this package.
- verbosity_level
Verbosity level. 0: No output, 1: Display steps, 2: Display progress in each step, 3: Display MCMC sampling.
- chains
Number of chains for Stan.
- iter
Number of iterations for Stan.
Examples
# \donttest{
data(d_sim_binom_cov_hgly2)
er_binary_cov_model <- dev_ermod_bin_cov_sel(
data = d_sim_binom_cov_hgly2,
var_resp = "AEFLAG",
var_exposure = "AUCss_1000",
var_cov_candidates = c(
"BAGE_10", "BWT_10", "BGLUC",
"BHBA1C_5", "RACE", "VISC"
)
)
#>
#> ── Step 1: Full reference model fit ──
#>
#> ── Step 2: Variable selection ──
#>
#> ℹ The variables selected were: AUCss_1000, BHBA1C_5, BGLUC
#>
#> ── Step 3: Final model fit ──
#>
#> ── Cov mod dev complete ──
#>
er_binary_cov_model
#> ── Binary ER model & covariate selection ───────────────────────────────────────
#> ℹ Use `plot_submod_performance()` to see variable selection performance
#> ℹ Use `plot_er()` with `marginal = TRUE` to visualize marginal ER curve
#>
#> ── Selected model ──
#>
#> stan_glm
#> family: binomial [logit]
#> formula: AEFLAG ~ AUCss_1000 + BHBA1C_5 + BGLUC
#> observations: 500
#> predictors: 4
#> ------
#> Median MAD_SD
#> (Intercept) -10.95 1.18
#> AUCss_1000 0.50 0.09
#> BHBA1C_5 0.50 0.09
#> BGLUC 0.74 0.14
#> ------
#> * For help interpreting the printed output see ?print.stanreg
#> * For info on the priors used see ?prior_summary.stanreg
# }
# \donttest{
data(d_sim_lin)
ermod_lin_cov_sel <- dev_ermod_lin_cov_sel(
data = d_sim_lin,
var_resp = "response",
var_exposure = "AUCss",
var_cov_candidates = c("BAGE", "SEX")
)
#>
#> ── Step 1: Full reference model fit ──
#>
#> ── Step 2: Variable selection ──
#>
#> ℹ The variables selected were: AUCss, BAGE
#>
#> ── Step 3: Final model fit ──
#>
#> ── Cov mod dev complete ──
#>
ermod_lin_cov_sel
#> ── Linear ER model & covariate selection ───────────────────────────────────────
#> ℹ Use `plot_submod_performance()` to see variable selection performance
#> ℹ Use `plot_er()` with `marginal = TRUE` to visualize marginal ER curve
#>
#> ── Selected model ──
#>
#> stan_glm
#> family: gaussian [identity]
#> formula: response ~ AUCss + BAGE
#> observations: 101
#> predictors: 3
#> ------
#> Median MAD_SD
#> (Intercept) 6.74 5.62
#> AUCss 0.46 0.04
#> BAGE 0.50 0.11
#> Auxiliary parameter(s):
#> Median MAD_SD
#> sigma 10.66 0.79
#> ------
#> * For help interpreting the printed output see ?print.stanreg
#> * For info on the priors used see ?prior_summary.stanreg
# }