A Bayesian Approach to Survival Analysis of Inverse Gaussian Model with Laplace Approximation
Year: 2016
Type: article
Abstract: The Inverse Gaussian distribution is a plausible model in settings where failure occurs when a deterioration process reaches a certain level. More generally, it is a reasonably flexible two-parameters family of models with properties that are rather similar to those of log-normal distribution. In this paper, an attempt has been made to outline how the Bayesian approach proceeds to fit such a model using Laplace Approximation, which requires the ability to maximize the joint log-posterior density. This model is applied to a real survival censored data, so that, all the concepts and computations will be around that data. R code has been developed and provided to implement in censored mechanism using analytic Laplace Approximation method. Moreover, parallel simulations tools are also implemented with an extensive use of R.
Language: English
Subfield: Statistics and Probability
Field: Mathematics
Domain: Physical Sciences
Open Access status: diamond
OpenAlex ID: W2574967310