Chi-square tests for generalized exponential distributions with censored data

Generalized exponential models have numerous applications particularly in reliability studies. Using the approach proposed by Bagdonavicius and Nikulin for censored data, we propose the construction of modified chi-square goodness-of-fit tests for the generalized exponentiated exponential model (GEE) and an accelerated failure time model with the generalized exponentiated exponential distribution as the baseline (AFT-GEE). Based on maximum likelihood estimators on initial data, these statistics recover the information lost while grouping data and follow chi-square distributions. The elements of the criteria tests are given explicitly. Numerical examples from simulated samples and real data have been presented to illustrate the feasibility of the proposed tests.