Bayesian and hierarchical Bayesian analysis of response - time data with itant variables.pdf.pdf

J. Biomedical Science and Engineering, 2010, 3, 711-718 JBiSEdoi:.37095 Published Online July 2010 (rnal/jbise/).Published Online July 2010 in and hierarchical Bayesian analysis of response - time data with itant variables Dinesh Kumar Department munity Medicine, Government Medical College, Chandigarh, India. Email:dinesh_******@Received 9 January 2010; 19 January 2010; 30 January 2010. ABSTRACT This paper considers the Bayes and hierarchical Bayes approaches for analyzing clinical data on re-sponse times with available values for one or more itant variables. Response times are assumed to follow simple exponential distributions, with a dif-ferent parameter for each patient. The analyses are carried out in case of progressive censoring assuming squared error loss function and gamma distribution as priors and hyperpriors. The possibilities of using the methodology in more general situations like dose- response modeling have also been explored. Bayesian estimators derived in this paper are applied to lung cancer data set with itant variables. Keywords:Bayes Estimator; Bayesian Posterior Density; Gamma Prior Density (GPD); Hierarchical Bayes Esti-mator; Hyperprior; Noninformative Prior Quasi-Density (NPQD); Progressive Censoring; Squared Error Loss Function (SELF); Whittaker Function W s1, s2 (.).