Method evidence record
Dynamic Bayesian Model Averaging
Dynamic Bayesian Model Averaging (DMA) extends standard Bayesian model averaging to settings where the best predictive model may change over time. It maintains a probability distribution over a set of competing models and updates that distribution sequentially as new observations arrive, allowing model weights to evolve rather than remaining fixed across the entire sample.
Source record
Citations copied verbatim from the method’s source record. No claim-level verification is inferred from them.
Dynamic Bayesian Model Averaging
Taxonomic method record · bayesian / bayesian
- Raftery, A. E., Karny, M., & Ettler, P. (2010). Online prediction under model uncertainty via dynamic model averaging: Application to a cold rolling mill. Technometrics, 52(1), 52-66. · DOI 10.1198/TECH.2009.08104
- Hoeting, J. A., Madigan, D., Raftery, A. E., & Volinsky, C. T. (1999). Bayesian model averaging: A tutorial. Statistical Science, 14(4), 382-401. · URL
Curated claims
Claims persisted in the evidence ledger, each with its own assessment.
No curated claims yet
This view does not invent a claim assessment when the ledger has none.
Related methods
Generated from the method graph and shown as machine-suggested relations — no evidence claim is inferred.