A Comparison Among Bayes Factors for Separated Models: Some Simulation Results
Maria Ivanilde Araújo, Basílio de Bragança Pereira
Abstract
Maria Ivanilde Araújo, Basílio de Bragança Pereira
Abstract
Some alternative Bayes Factors: Intrinsic, Posterior and Fractional have been proposed to overcome the difficulties presented when prior information is weak and improper prior are used. Additional difficulties also appear when the models are separated or nonnested. This paper presents some simulation results comparing these Bayes Factors to discriminate among the Lognormal, Weibull, Gamma and exponential distributions.
OpenAlex reports 1 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
Some alternative Bayes Factors: Intrinsic, Posterior and Fractional have been proposed to overcome the difficulties presented when prior information is weak and improper prior are used. Additional difficulties also appear when the models are separated or nonnested. This paper presents some simulation results comparing these Bayes Factors to discriminate among the Lognormal, Weibull, Gamma and exponential distributions.
Key concepts: Weibull distribution, Bayes' theorem, Bayes factor, Statistics, Log-normal distribution, Econometrics, Mathematics, Exponential function