Data driven conclusion is mostly accepted approach in any medical research problem. In case of limited knowledge of deep idea about supportive data on the problem, automatic digging of the variable plays important role for insight view of the study. Bayesian model averaging can be considered for automatics variable selection. It can be used as an alternative of stepwise regression method. The aim of this paper is to show the application of Bayesian modeling averaging in medical research particularly in cancer trial. Method is illustrated on Bone marrow transplant data. It can be recommended that BMA can be used frequently in data selection and as a tool of exploratory data analysis method. It is very handy method of choice for data analysis.
Bhattacharjee, Atanu, "Bayesian Model Averaging:- An Application in Cancer Clinical Trial" (February 2014). COBRA Preprint Series. Working Paper 106.