Skip to contents
mlumr 0.1.0
Initial CRAN Release
- Implements Multilevel Unanchored Meta-Regression (ML-UMR) for population-adjusted indirect comparisons in disconnected networks
- Supports three outcome types: binomial, normal, and Poisson
- Two model variants: SPFA (shared prognostic factors) and relaxed SPFA (treatment-specific coefficients)
- Quasi-Monte Carlo integration using Sobol sequences with Gaussian copula for aggregate data
- Bayesian inference via Stan/cmdstanr with HMC sampling
- Comprehensive diagnostics: DIC for model comparison, convergence metrics (Rhat, ESS), posterior predictive checks
- Publication-ready visualizations with ggplot2
- Simulation data generation for testing and validation
