Imprecise inference based on the log-rank test for accelerated life testing

Abstract This paper presents an imprecise predictive inference method for accelerated life testing. The is largely nonparametric, with a basic parametric function to link different stress levels. log-rank test used provide imprecision the parameter, which in turn provides robustness resulting lower and upper survival functions future observation at normal level. An application using data from literature presented, simulations show performance of method. In case model misspecification, may be achieved price large imprecision, would emphasize need more or further assumptions.