A0481
Title: Accelerated failure time model based on nonparametric Gaussian scale mixtures
Authors: Byungtae Seo - Sungkyunkwan University (Korea, South) [presenting]
Abstract: When some parametric error distributions are assumed, such as normal for the accelerated failure time model, estimators typically suffer from misspecification problems. To relax this problem, a nonparametric Gaussian scale mixture model is proposed to flexibly specify the error distribution. Unlike existing nonparametric or semiparametric estimation methods such as rank based procedures, the proposed method enables us to use an explicit likelihood function while avoiding potential misspecification problems. This model is presented with specific estimating algorithms and some numerical examples.
