Home Page

Case Study

Bayesian Univariate Logistic Regression (Laplace + MH)

2025
  • R
  • Statistical Computing
  • Bayesian Methods
  • Parallel Computing

Posterior mode via Newton–Raphson; Laplace approximation; MH sampler; parallelized 60 fits with snow; posterior means + MLE sanity checks.

Problem & Motivation:

Approximate posteriors quickly and validate with sampling.

Data & Approach:

  • Mode finding; Laplace evidence; MH initialized at mode; log-accept tracking; parallel runs.

Results:

  • Laplace close to MH posterior means; acceptance stable with tuned proposals.

Limitations:

Univariate only; Gaussian priors.