Diagnostics for a class of survival regression models with heavy-tailed errors

The heavy-tailed survival regression models provide a useful extension of the normal regression models for data sets involving errors with longer-than-normal tails. This article develops influence diagnostics under case-deletion model (CDM) in survival regression models for which the errors follow the log-generalized Birnbaum–Saunders distribution based on the t model (LBST). The one-step approximations of the estimates in CDM are given and case-deletion measures are obtained. Meanwhile, we discuss a score test for the homogeneity of shape parameter in LBST regression models. One numerical example is given to illustrate our methodology and the properties of the score test statistic are investigated through Monte Carlo simulations under different censoring percentages.