Minimum message length inference of the Poisson and geometric models using heavy-tailed prior distributions
Autor: | Chi Kuen Wong, Daniel F. Schmidt, Enes Makalic |
---|---|
Rok vydání: | 2018 |
Předmět: |
FOS: Computer and information sciences
Estimation theory Applied Mathematics Model selection 05 social sciences Bayesian probability Poisson distribution Bayesian inference 01 natural sciences 050105 experimental psychology Minimum message length Methodology (stat.ME) 010104 statistics & probability symbols.namesake Prior probability symbols 0501 psychology and cognitive sciences 0101 mathematics Minimum description length Algorithm Statistics - Methodology General Psychology Mathematics |
Zdroj: | Journal of Mathematical Psychology. 83:1-11 |
ISSN: | 0022-2496 |
Popis: | Minimum message length is a general Bayesian principle for model selection and parameter estimation that is based on information theory. This paper applies the minimum message length principle to a small-sample model selection problem involving Poisson and geometric data models. Since MML is a Bayesian principle, it requires prior distributions for all model parameters. We introduce three candidate prior distributions for the unknown model parameters with both light- and heavy-tails. The performance of the MML methods is compared with objective Bayesian inference and minimum description length techniques based on the normalized maximum likelihood code. Simulations show that our MML approach with a heavy-tail prior distribution provides an excellent performance in all tests. |
Databáze: | OpenAIRE |
Externí odkaz: |