Title: Genetically modified mode jumping MCMC approach for Bayesian multivariate fractional polynomials
Authors: Aliaksandr Hubin - NMBU (Norway) [presenting]
Riccardo De Bin - University of Oslo (Norway)
Georg Heinze - Medical University of Vienna (Austria)
Abstract: A framework is suggested to fit fractional polynomials based on the Bayesian Generalized Nonlinear Models. A version of the Genetically Modified Mode Jumping Markov Chain Monte Carlo (GMJMCMC) algorithm is adopted. Preliminary simulation runs show promising results in terms of identifying the data generation mechanism: The suggested approach uniformly outperforms the existing Bayesian fractional polynomial framework oth in terms of Power and false discovery rate (FDR). Also, the performance is on par (somewhat better) with that of frequentist fractional polynomials. Still, the results indicate that work on the priors is likely to improve the performance even further.