bmbeR: Bayesian Model Building and Evaluation Repository

This repository contains a set of R scripts designed to build, evaluate, visualize, and perform sensitivity analysis on Bayesian models. These scripts use a mixture of rstan, rstanarm, and other Bayesian analysis libraries to facilitate the modeling process.

Scripts and Their Functions:

01_libraries_and_setup.R

• Purpose: Loads the necessary libraries for the entire modeling workflow.
• Libraries Used:
  • rstanarm
  • bayesplot
  • ggplot2
  • cowplot
  • purrr
  • rstan
  • loo

02_distributions.R

• Purpose: Defines various prior distributions.
• Functions:
  • student_t_prior
  • normal_prior
  • cauchy_prior
  • uniform_prior
  • beta_prior
  • gamma_prior
  • binomial_prior
  • poisson_prior
  • lognormal_prior
  • bernoulli_prior

03_utilities.R

• Purpose: Provides utility functions to manage distributions.
• Functions:
  • add_distribution: Add a new distribution to the working list.
  • reset_distributions: Resets custom distributions to their original forms.
  • get_prior_distribution: Retrieve a prior distribution based on type.

04_empirical_bayes.R

• Purpose: Generate empirical Bayesian priors from data.
• Functions:
  • empirical_bayes_priors: Computes priors based on data and given formula.

05_model_convergence.R

• Purpose: Checks the convergence of a given model.
• Functions:
  • check_convergence: Assesses if a model has converged.

06_model_fitting.R

• Purpose: Fits a Bayesian model using given priors and data.
• Functions:
  • fit_model_with_prior: Fits a Bayesian model using stan_glm.

07_model_visualization.R

• Purpose: Visualizes model fit and posterior distributions.
• Functions:
  • generate_plot: Generates trace or histogram plots.
  • plot_posterior_distributions: Plots 95% intervals for posterior distributions.

08_model_sensitivity.R

• Purpose: Performs sensitivity analysis for various prior configurations.
• Functions:
  • define_prior: Defines priors for the sensitivity analysis.
  • sensitivity_analysis: Conducts sensitivity analysis using different prior combinations.

09_model_evaluation.R

• Purpose: Evaluates the performance of a model on test data.
• Functions:
  • evaluate_model_performance: Computes RMSE for regression or accuracy for classification.