Package index • mlumr

Data setup

Assemble the index IPD and comparator AgD into one mlumr_data object and build the quasi-Monte Carlo integration points for population adjustment.

set_ipd(): Set up individual patient data (IPD)
set_agd(): Set up aggregate data (AgD)
set_agd_surv(): Set up aggregate survival data (reconstructed pseudo-IPD)
combine_data(): Combine IPD and AgD for unanchored comparison
add_integration(): Add numerical integration points
check_integration(): Check integration point adequacy
check_identification(): Can the aggregate data identify the comparator coefficients?
unnest_integration(): Expand integration points into a long-format data frame
make_knots(): Choose M-spline knots for a flexible-baseline survival model

Marginal-distribution helpers for the comparator integration points.

distr(): Specify a marginal distribution
qbern(): Bernoulli quantile function
pbern(): Bernoulli CDF
dbern(): Bernoulli PMF
qgamma() pgamma() dgamma(): The Gamma distribution, parameterized by mean and standard deviation
dlogitnorm() plogitnorm() qlogitnorm(): The logit-Normal distribution

Fit the Bayesian ML-UMR model (SPFA / relaxed) and select the Stan backend.

Prior constructors, the prior summary, and prior-sensitivity analysis.

prior_normal(): Specify a normal prior
prior_student_t(): Specify a Student-t prior
prior_cauchy(): Specify a Cauchy prior
prior_exponential(): Specify an exponential prior
prior_summary(): Summary of priors used by a fitted ML-UMR model
prior_sensitivity(): Prior sensitivity analysis for an ML-UMR fit
default_prior_intercept() default_prior_beta() default_prior_sigma() default_prior_aux() default_prior_smooth(): Default priors used by mlumr()

Population-standardized effects and absolute predictions in either population.

Unadjusted (naive) and simulated-treatment-comparison (STC) reference estimators.

Compare SPFA against relaxed, and the information criteria behind it.

The plot() methods, the Kaplan-Meier overlay, the method-comparison forest, and the prior-versus-posterior overlay.

plot(<mlumr_prediction>): Plot absolute predictions from a fitted ML-UMR model
plot(<mlumr_marginal_effects>): Forest plot of population-standardized marginal effects
plot(<mlumr_conditional_effects>): Plot covariate-conditional treatment effects
geom_km(): Observed Kaplan-Meier layer for survival overlays
mlumr_forest(): Forest plot of a small set of estimates
plot_prior_posterior(): Prior-versus-posterior overlay

Bundled index IPD and comparator AgD for the worked examples, one pair per outcome family (binary, continuous, count, survival).

psoriasis_ipd: Plaque psoriasis: index individual patient data (binary)
psoriasis_agd: Plaque psoriasis: comparator aggregate data (binary)
shoulder_ipd: Shoulder pain: simulated index IPD (continuous)
shoulder_agd: Shoulder pain: simulated comparator aggregate data (continuous)
caries_ipd: Dental caries: simulated index IPD (count)
caries_agd: Dental caries: simulated comparator aggregate data (count)
ndmm_ipd: Newly diagnosed multiple myeloma: index individual patient data (survival)
ndmm_agd: Newly diagnosed multiple myeloma: comparator pseudo-IPD (survival)
ndmm_agd_covs: Newly diagnosed multiple myeloma: comparator covariate summaries (survival)