On Application of Bayesian Mixture of Linear Mixed-Effects Models to MLB Player Salaries

  • 謝 岱凌

Student thesis: Master's Thesis


We consider Bayesian variable approaches to simultaneous selection of important fixed and random effects in the finite mixture of linear mixed-effects models Latent variables are introduced to classify the membership of observations and to facilitate the identification of influential fixed and random components in the longitudinal data A spike-and-slab prior for the regression coefficients is adopted to sidestep the potential complications of highly collinear covariates and to handle p>n in the variable selection problems We employ Markov chain Monte Carlo (MCMC) sampling techniques for posterior inferences and explore the performance of the proposed model on simulated data Two actual datasets MLB salary data and psychiatric data are used to explain the difficulties and limitations of the proposed model in real applications
Date of Award2017 Aug 3
Original languageEnglish
SupervisorKuo-Jung Lee (Supervisor)

Cite this